I did the research too, some said the Sigma was slightly better at 400mm, the Tamron was reputedly slightly sharper in the centre of the image but the Sigma was more consistent across the full frame. Some maintained the AF was better at tracking fast moving objects on the Tamron (using a DSLR), but I believe that was before the Sigma got the firmware update, similar story with the OS (image stabilisation). As mentioned previously, the Sigma zoom turns the same way as Canon zooms, the Tamron turns the opposite way, which you may find annoying. The Sigma also allegedly offers more 'tweakability' with its Sigma dock system, including a 4 position fine tuning facility for the focus. The Tamron has full weatherproofing where the Sigma has partial weatherproofing, but I don't take my kit out in the rain or use it in dusty conditions, so weatherproofing wasn't a deciding factor for me. However, I believe the Tamron is more expensive than the Sigma, unless there's an offer on somewhere.
All in all, I don't think you can probably put a cigarette paper between them in terms of real world use for landscape and casual wildlife (rather than lightning fast AF and tracking, where the MkII Canon 100-400L should really shine). With such apparently minor optical differences between the Sig and Tam, any 'advantage' one allegedly has over the other may well vary with individual copies of each mass-produced lens. Results from a well-tuned and accurately focussing lens should always look better than one that's slightly off straight out of the box.
I also like the colour and contrast of photos taken with my Sigma lens, which might be a factor if you compare some 'unprocessed' shots from each lens when used on a Canon - a Flickr search might help there, but as we know, not everyone uploads full-size, unaltered photos to Flickr, and you need to compare like with like. So there's another hour or so of your life that you won't get back looking at that lot!