Memrise sure seems to have a very unique definition of a “bug.”
Everyone else defines a bug as an unanticipated and undesired behavior of a program, but in this case, the mems are displaying exactly as specified in the stylesheet that their designer wrote. They constrained the mems to the interior of a container with a small fixed width of 435pixels, and 75pixels of side padding. That leaves just 285 pixels of width for the mem to appear in, which is exactly what they were doing. So how is that a bug? It is entirely predictable, and in exact conformance with what their designer wrote. This was the controlling style yesterday:
Today they’ve changed he container size a bit, to 510pixels, so an increase of just 75 pixels in width.
So I don’t understand how they think that they can legitimately call it a bug, when we can read the CSS, and see that it was intentional and it behaved exactly as specified. Obviously it was a bad design decision, but that doesn’t make it a “bug.”
From the discussion I had, this was not intentional. There were no direct changes towards mems itself.
We upgraded from bootstrap 2 to 3 which changed the box model. The box sizing changed to border-box in that library as a result we have had to change the dimension of some of the boxes.
The relevant CSS style is not a part of the standard bootstrap code, but rather custom Memrise code.
The width and padding of the container isn’t a random, unanticipated side effect of a change in the bootstrap code or its box model. The Memrise designer manually specified those fixed dimensions precisely, and they could see in advance exactly how the mems would look with that style applied to them, before they published their code. They and their supervisor must have been satisfied with the result, because nothing about this was unpredictable or a mysterious side effect.