My suggestion is not about ranking, just selecting the Top 10 by round. While the first results of 10 sites will likely be some of the most impressive, I see the Top 50 as more or less a list of equals. Each member would select (10) sites per round, nothing ranked. Whichever sites have the most votes make it on the list. After round 3 (Top 30), we could perhaps review the results and decide about the overall number for "Top 50" missing. It does not have to be 50, but since the original was 50 it seems like generally this is our target. Moreover, I feel that through voting we can select a strong Top 50, I don't (at this point) feel it will be full of non-deserving sites.
This approach promotes honest voting, for example there are key sites that I will vote for in round 1 & 2 that are the absolute most deserving (widely viewed as Top 50), and not necessarily sites that I originally suggested. I will have a chance to pick some of my personal favorites in later rounds. Since each member has plenty of opportunity to participate (vote) they should not feel obligated to select national or personal selections in the early rounds. We are yet to see how this might effect the later rounds, will there be any kind of consensus, who knows, that's why we vote!
I wouldn't mind trimming the list a bit, indeed I promoted this. But the consensus was this was not necessary. While I prefer trimming, ultimately I do not think it makes a huge difference.
As far as the Top 50 missing list being "final", I don't view it this way exactly. It will be nice to have a more complete, agreed list. Yet, I assume that we can revisit the list every year when sites are inscribed. Our original list only had a couple inscriptions, but I feel our new agreed list will likely see more inscriptions.
I think we might have to concede that whatever we create here will not be perfect, sites with only a couple votes (2-3) may make it on the list in the later rounds. But we should also recognize this approach is far superior to the original Top 50 list which has about 10-20 sites that had no consensus.
I REALLY feel that voting on the sites we just went through the process of selecting would be a step backward, and would ultimately decrease the likelihood of a more balanced/deserving group of sites. For example, if the 50 odd sites that were part of everyone's 3 favorites were voted upon, how many European & American sites would stay, how much other sites would not receive 3 votes? This is inevitable based on the majority of this forum being from Europe, no? We got a couple suggestions from non-Europeans, but I fear these would be the first to not receive 3 votes. This does not necessarily mean they are not deserving, but generally a lot of us might not have much knowledge of Indian or Philippine nominations. Yet, if they are part of the voting rounds, they might have a more realistic chance in the later rounds.