The context that this article misses is that PDF is a
huge spec. It is much bigger than you think it is. Adobe maintains the reference implementation. How many image formats do they have to maintain? How many compression schemes?
It is not as if Adobe can simply set a new code standard and fix all that old code instantly.
Adobe's legacy code problem with PDF's is worse than Microsoft's; Adobe has to maintain some of this gross old stuff for all time.
