Few beginners find learning to program easy. There are many factors at work in this phenomenon with some being simply inherent in the subject itself, while others have more to do with deficiencies in learning methods and resources. As a result, many programming environments, software applications, and learning tools have been developed to address the difficulties faced by novice programmers. Of these tools, visual-based tools and the use of visualization have proven to be very effective in helping novices overcome several of these traditional difficulties. In this paper, we first examine the traditional difficulties that novice programmers encounter when take an introductory-level programming course are examined. It is important to gain an understanding of the scope of these difficulties first, as the rest of this paper considers how visual tools, visualization, and UML can be utilized to aid novice developers in these areas of difficulties. Next, we provide an analysis of several modern visual learning tools, including EROSI, AnimPascal, BlueJ, FLINT, BOOST, and SOLVEIT. In particular, we look at how these tools use visualization to help mitigate the difficulties novice programmers face. Each tool is also assessed based on its overall effectiveness of using visual aids and visualization to help the beginning programmer. We then turn our attention to the Unified Modeling Language (UML) and how it can be utilized to help the novice programmer in system design and modeling. The UML specification is carefully discussed, and aspects of the specification that hold the most potential for aiding novice programmers are identified. Finally, we focus on UML modeling and present the theoretical foundation for a new visual learning tool based on the UML standard. This proposed learning environment attempts to combine promising attributes of existing tools we previously examined, along with the potential benefits of UML-based modeling. The proposed tool would provide a superior learning environment for the novice programmer for several reasons. First, it is heavily based in the visual domain. Visual tools have continually proven to be extremely powerful in helping novices in learning abstract computer concepts. In addition, visualization helps novices construct a mental model of concepts, which is pivotal to further comprehension and understanding. Second, the proposed environment would allow for a constructivist learning approach, constraining the UML domain for novices, yet easily expanded for more complex projects as the student progresses. Finally, this tool would naturally aid in solution delivery and documentation of the learner's path to solution. (Contains 9 figures and 2 tables.)