What is the green cube that shows up when I highlight my name on the overview page? I assume the blue cubes are the ones I have completed.
Are the empty blue cubes on the overview page areas that I have done but no one else has?
When I encounter an area I am unsure about should I be conservative and stop or blaze on? In the event that I do continue on and I turn out to be incorrect, is the computer able to utilize the area I have filled in if it corresponds to another task?
During some tasks the images seem to skip a slide, what causes that?
The green cubes indicate the locations that online users are playing right at the moment. The blue cubes are, as you figured, the locations that you have completed so far (In fact blue cubes appear when your ID is clicked on and green when not).
To stop or to blaze, well, I couldn’t find any better answer; just follow your best guess. That being said, I find that most users miss branches more frequently than they add wrong branches (Some people do add wrong branches when they are novice, but they tend not to after they step forward).
We are not recycling the additional pieces for the moment, but they are being saved in the database. I believe we could find a right way to utilize them in near future.
There are jumps between slices. Some are small, some are big. They are defects in the image stack. Please refer previous discussions:
When the images are “stitched” together, the computer appears to have difficulty seeing that the images connect sometimes. Is there a way to create alternate tasks where there is a slight shift in the imagery for the computer to cross reference?
@krittenhouse This issue actually has to do with how browsers display images. All the images are compressed as JPEG which is a lossy form of compression so it creates some defects in the images called artifacts. In order to adjust for this, your browser does what’s called filtering of the images. It tries to smooth them out so that they look more natural and you can’t see the artifacts so much. Turns out that they are great at filtering individual images, but not so great at filtering across images which are placed next to each other. It’s something that we’ve tried to minimize the effect of, but you can still see it occasionally. If the browser wasn’t filtering or if we were sending you uncompressed images then you wouldn’t even be able to see the seems where they are “stitched” together. That having been said there are problems with both of those as well. We think that a few uneven seams is the best tradeoff we can make in this situation.
Are there four images placed together in the tasks?
Ah, were you talking about between different slices? So what I was talking about is a single slice is comprised of 4 images. If you zoom in, you can probably see the boundaries that I was talking about.
Aligning things between slices is based on a process called registration & alignment. We try to find points that correspond between different slices, and we line them up so that they overlap with each other. We do it in such a way as to minimize the errors in alignment. Unfortunately things can happen which make this process imperfect. Different slices can be distorted. (We call them slices because they are actually different physical slices of tissue). Slices can be missing or or otherwise damaged. The moral of the story is that while we do the best that we can to line things up correctly, it will never be perfect, no matter how much we shift them in different directions. The good news is that it works almost perfectly the majority of the time.
How hard can it be to just slice of a perfect chunk of tissue a few nano-meters at a time? ;) Want me to help, I have really steady hands!