Okay, this is a long one...
I strongly disagree with the idea from earlier in the thread that panels should be composed entirely of non-members of the BFS, and such a proposal would, I think, be very unlikely to get through an AGM in any case, although I've been surprised before (e.g. when they gave the go-ahead for ebook-only memberships last year). You might as well send them off to be decided by Martians - they're the British Fantasy Society's awards, and they should be chosen by our members (although Johnny's idea about past winners is interesting - there's an argument that past winners should at the least be given the permanent right to vote in our awards).
We don't need to go out looking for professional readers; we just need to identify the members willing to take on the responsibility. It might not be a bad idea to have one outsider on each panel, someone the rest of the panel has to look in the eye and say, yes, this really is the best book on this list. But panellists have to do that with each other anyway. The Best Newcomer panel has been made up of members for the last two years, and there's been no complaint whatsoever about their choices.
When the BFS tried having panels for all the awards in the seventies they were published in Prism - a good idea, I think. It means the panellists, if they make a really unexpected choice, know that their decision and motives will come under scrutiny, and that forces you to take the responsibility seriously, something that a significant proportion of voters aren't really doing. I saw one guy on Facebook saying that the grumbling about the results was tantamount to defamation, before saying at the end that he just voted for the people he knew... The problem is that the more principled and ethical the member, the more likely they are to say, I'm not voting because I hadn't read the books.
Anonymity: I'm not sure I agree either that anonymity in voting is important or desirable. Only the awards admin should know who is voting for who, and you can't check if people are recommending their own work without knowing who they are. You have to be able to check that people were members at the time when they voted. Also, the awards administrator being able to spot certain patterns in the voting can help them to identify which rules need tweaking. The problem this year hasn't been with people wanting to gain favour with those in charge, but more, as far as I can see, about a group of members with a shared interest in seeing particular people win. For one thing, the membership wasn't even aware, as far as I know, who was actually running the awards until the AGM.
That is, of course, a problem in itself: the BFS *must* announce immediately when a position has changed hands, even temporarily, so that any ethical questions can be raised before the event, rather than after. I appreciate and agree with David's point that with the work involved, it would have been impractical to step down in the middle of the awards. I'd suggest, though, that (i) the handover should have been made public, giving members the chance to express any concerns, and (ii) it would have been wise - either at the point he started to help out, or at the point when he wholly took over - to withdraw Telos from the category that might earn him a cash prize.
Banning the work of everyone who is working or has ever worked on the committee from receiving awards is impractical and undesirable, I think. Let's say author X's forthcoming novel from Quercus is such a corker that it's up for an award next year - it would be daft to exclude it on the basis that it's edited by Jo Fletcher and she was on the BFS committee umpteen years ago. The BFS is essentially a small press - it needs more people on the committee with publishing experience, not fewer, so let's not discourage them from volunteering.
Yes, the BFS has in the past rewarded people like Ramsey Campbell, Graham Joyce and Stephen Jones with multiple awards, and that's because they are the people producing the kind of work that BFS members tend to love. They were winners that, while reflecting the society's inclination towards weird fantasy over the heroic variety, FantasyCon attendees were happy to see win prizes. That insularity - which you also can see in calls to abolish the awards for film and television on the grounds that the winners don't come to FantasyCon - is a slightly different issue from the kind of boosterism that is so evident, for example, in those daisy chains of mutual five star reviews on Goodreads, and which to all appearances has produced this year's curious results.
I hope that the reaction to this year's results will have a positive effect even if the rules don't change, because it will encourage people to think twice before helping to push someone they are friendly with into a potentially awkward and embarrassing position. The results of these awards come under intense scrutiny every year, and if the winning material struggles to stand up to that scrutiny - whether because it's not very good, or because it's, you know, Sherlock! - questions are always going to be asked about how it came to win.
One last thing for now: it's been suggested somewhere that the awards administrator be a complete outsider. I doubt that would work, unless the BFS was willing to pay someone. I would however seriously suggest that the awards administrator not take part in BFS committee discussions. The BFS committee can be a tumultuous place (probably less so since I left, admittedly). Awards admin is a position that benefits from stability, consistency and experience: you need to know the rules inside out. Give them a free BFS membership for their work, of course, and let them lead committee discussions on rule changes, but keeping the awards admin out of the political side of things would, I think, have long-term benefits.
Okay, one more thing: I'm very impressed that David is sticking it out. No one who saw his immensely moving speech at last year's awards could have the slightest doubt about his love for the British Fantasy Society and his respect for the awards. While I'm disappointed by the results, I have no doubt that they would have been precisely the same if I had still been awards administrator.