On the whole I am impressed but not entirely convinced by the Gettier criticism of justified true belief (JTB) as a good definition for knowledge. It seems possible to define the justification criteria to exclude Gettier cases as being true knowledge, and that feels intuitively the right approach to take. I can't believe it is necessary to add lots of additional caveats or criteria to exclude Gettier cases from qualifying as knowledge, instead we need to be clearer about what we mean by justification.
I think perhaps Gettier examples are "reasonable assumptions" but not "knowledge".
So to be justified, knowledge needs to be based on reasoning that takes account of chance, luck, misapprehension, illusions etc. This is why Descartes came to the conclusion that there was only one bit of knowledge of which he could be entirely sure - the fact that he existed!
Of course the amount of justification required depends on the circumstances and the use to which the knowledge is being put. If someone asks you the time because the want to catch a train, the justification needed is less conclusive than if they need the time to calculate the longitude of a ship crossing the ocean.
So in conclusion there are certainly different standards by which we would assess propositions as being knowledge, and these standards vary significantly depending on the circumstances in which the statements are made.