Did you click the link? The page I submitted contained an additinonal video demonstration which I found far more interesting than the original source PDF.
Additionally, I didn't think it was appropriate to directly link to a PDF download. Generally, readers don't expect that.
Thanks for finding that. I literally searched the source of the blog article for the video src url and then searched the PDF hoping to find some match pointing to a more authentic source.
Yeah, I knew the MSR page and that you have to basically start at their front page. Because of course you can't just remove the file from the paper URL and get an overview or something like that, that would be to easy...
It seems to me like the internals of their content management system are showing through and making it complicated. "Oh, a video? That's something completely unrelated and a stand-alone artifact"
FWIW, I think submitting papers with a [PDF] in the title is fine, because many people seem to start in the comments anyways, but that's just my opinion.