Scale is everywhere in all geosciences, with geographic information science being unexceptional. It has been 30 years since Abler published an article in 1987 which lists scale as one of several key research topics in geographic information science. Over the past 30 years, a lot of researches on this topic has been carried out by researchers from different countries. It seems to be appropriate time to have a close examination into the issues of this topic. This paper attempts to make a brief review of these researches. The discussions will be focused on the five key issues raised by Goodchild and his collaborators, i.e., scale invariants, ability to change scale, measure of the impact of scale, scale as a parameter in process models and implementation of multiscale approaches. It has been found that the impact of scale has been well studied, but the joint effect of scales at multiple stages (e.g., data scale and analysis scale) is still poorly investigated; various transformation models have become available, but the theoretical basis of some methods is not sound enough; the issue of "scale as a parameter in process model" has a long way to go. It is also noted that the effect of scale on information is rarely touched. All these issues deserve great attention from geo-information community.