The large increase in the number of Web accessible data sources has raised new challenges such as (a) identifying relevant data sources from the available ones and (b) answering user’s queries according to their requirements. Traditional integration systems are facing problems because (i) sources are autonomous and semantically heterogeneous; (ii) their database design and schemas are different and (iii) their query languages are also different. In this paper the authors have highlighted some issues of semantic-based query rewriting and propose a solution to (a) provide uniform access to all sources.