I was catching up on this thread, with no intention of dragging the silliness out, but I got caught by your math and made the light hearted comment, didn't bother to check the actual position of the post relative to the page count.
If anything I know someone will drag out that number when this topic rears it's ugly head again, so may as well call it out now before this 0.5% or whatever number gets trotted out as fact in the future.
First I would say that only Bill would have the data to determine "what percentage of TMP'ers don't like Tango", and I personally feel he should ignore any such request.
But if you were to do such an exercise the methodology would be:
1. Determine TMP audience size
- Get all total registrations, then
- De duplicate the data set, so you only count people once. We know that there are a lot of duplicate accounts here, heck even I have an old one that predates 2008. The point is you only should count someone once and not multiple times.
- Remove all banned and locked accounts, not point counting them is there.
- Remove all "dead" accounts, i.e. memberships that haven't been used to access the site in the past six months. You can go twelve, but either way if someone hasn't used a website in that period then the chances of them coming back are slim – it does happen but not in a statistically significant way. Once you have this number it's on to step 2.
2. Now you can filter this result by activity to get a picture of the active user base. You'd need to agree to definitions from here on in, and there will be many different interpretations of what an active user is. One suggested definition would be -
- Has visited TMP at least once a week within X period, where X would be (in this instance) since Tango joined our community.
Once you filter down to this number, you'll have a pretty good idea as to the size of TMP. There are a fair few other filters that web publishers would use to get a much more accurate number, especially as you can split the numbers by "active" and "passive" memberships. If we can define active as visits at least once a week, you can see what "passive" would be.
But the main point is that there isn't 36+k members, there have been 36+k membership registrations. Very different numbers. Using your methodology is like saying that Ronald Reagan never enjoyed majority support in the USA by comparing his number of votes from an election to the entire population of the USA since 1776.
I'm certainly not saying that those who have an issue with Tango are in the majority, far from it and frankly we don't need to know, but I am saying that methodology you came up with is 100% inaccurate.
