Dr Shiva brought up some interesting questions and I think we definitely need good answers for them from the county. And it also shows that just because there isn’t obvious evidence of fraud, doesn’t mean evidence can’t be found when you actually start investigating.
However one curious thing about his report is the question he had about how the stamp appears to go under the triangles that were printed on the envelope. He implied there was fraud involved here and that it might be some sort of “photoshop mistake”. The guy has 4 degrees from MIT, some related to computer science, it sounds like he has extensive experience in computer vision, so how is it possible that he (nor any of his team) realize they were looking at not raw images, but the output of an edge detection filter, which plainly explains why the text appears to be “under” the triangles?
It’s not possible that he could have missed a detail that anyone who paid attention in the first week of an intro computer vision class would have fought. That sort of implies his whole analysis was not done in good faith and should be thoroughly scrutinized before any other of his implied conclusions are to be taken seriously.
> Dr Shiva brought up some interesting questions and I think we definitely need good answers for them from the county.
At least some purported answers can be found on Garret Archer's twitter [0]. If the answers there are accurate, it goes a long way to show that either lots of communication or intimate knowledge of the ins and outs of the election process is very important for tracking down every last discrepancy.
However one curious thing about his report is the question he had about how the stamp appears to go under the triangles that were printed on the envelope. He implied there was fraud involved here and that it might be some sort of “photoshop mistake”. The guy has 4 degrees from MIT, some related to computer science, it sounds like he has extensive experience in computer vision, so how is it possible that he (nor any of his team) realize they were looking at not raw images, but the output of an edge detection filter, which plainly explains why the text appears to be “under” the triangles?
It’s not possible that he could have missed a detail that anyone who paid attention in the first week of an intro computer vision class would have fought. That sort of implies his whole analysis was not done in good faith and should be thoroughly scrutinized before any other of his implied conclusions are to be taken seriously.