A couple of comments from me
First of all - while I am not involved in OA from day to day, the original OA devs have moved to Emsisoft, and added to this is Fabian and the Emsisoft team. To take over an application like OA is no small task, and as I can see the reworkings of OA are actually rather good. OA is still, compared to the big end of town, a small fish in a big pond and it has been forever reliant on mutual goodwill between the OA team and the community.
The whitelist - I used to spend an age maintaining it - but as Fabian demonstrated - it's a HUGE database, and a huge job. From one side, something like VLC could have many different exes, many different DLLS, and many different versions - in my install there are 3 executables, and 272 dlls.
Each one of these used to have to be manually trusted in OASIS. For each version. So if VLC did a new build, all the hashes can change and hey presto another 270+ files appeared in OASIS that need to be validated manually. Consider then then top 100 applications, and you can see that there's a lot of work.
This was mitigated a little by use of digital signatures comparison; but even then, someone has to do it.
So - if you believe in whitelisting, clearly there has to be a better way.
Just for fun, one time I did a quick calculation at the number of unique files in OASIS. I worked out that if I got the entire beta test team taking 5 minutes per file to perform an assessment (there were about 20 or so beta members), working 15 hours per day, it would take something like 6 months to process all of the OASIS files, assuming, of course, no new ones were added. (these numbers are "made up" I don't recall the actuals, but you get the idea).
Whitelisting or not is something Emsi have to decide they want to continue, or to develop white heuristics, or other ways of determining safeness, or protecting against dangerousness. Either way will take time and resources.
So, I don't think Emsi has wrecked OA - from the changes I saw in last release there were some very welcome and long overdue improvements to the product, as well as removing some obsolete features.