
I think that supporting a platform should solely depend on how large it's market share is overall - across all platforms - even if it does not support all of them. While a browser's cross-platform support is generally desirable, I personally think that this won't matter as much as the market share:
i.e. Opera may be cross platform but it's market share is so low that even if you combine their user base over all platforms it would be smaller than Safari's user base. Hence developing for Safari should be more desirable than developing for Opera as you would reach out to millions more of users which can support the project with donations. (I have no idea what your plans on supporting other browsers are, so this paragraph is just hypothetical.)
The stat's from findthebest.com are new to me because in the report that I usually read the Internet Explorer has the largest market share. Of course such stats depend on how you gather the information. Browser web traffic is not directly proportional to the user base - I assume. But I think discussing browser market share it a topic for another thread...

I have a problem following your analysis of findthebest's graphs. Where can you see that out of the 7% only 5.92% of users choose Safari? The 5.92% is in relation to all the users of all platforms as far as I understand. Therefore, I think that it means that overall 5.92% of the users use Safari. That would leave 1.08% for Chrome on Mac, Firefox on Mac,... .
Personally, I like to read the ArsTechnia article about browser market share: http://arstechnica.com/information-tech ... stay-flat/. They retrieve their sources from netmarketshare.com. According to these stats, Safari is among the four top dominant browsers for desktops. They say that overall the desktop browser market share of Safari is 5.26% which is similar to the 5.92% that findthebest mentioned. After Chrome and Firefox, Safari would be the best next option to support. Here are ArsTechnica's stats:
