Visual programming languages are object-oriented in nature, hence, extra codes have to be created to accommodate operator overloadings in the modules. These extra codes slow down the execution time of the programs especially when the programs are big.
Visual programming languages are usually designed and targeted for certain purposes. Most of the languages that our group has explored so far are mostly good for number crunching (except KidSim, where it is designed for kid to play around with their newly invented games). These languages are limited in functions or modules to let programmers program our everyday use programs. Further more, programs written in visual programming language can not generally be compiled. You can only compile the program if the visual programming languages can produce C or other languages equivalent codes.
Due to the complexity of large program, it is hard to visualize everything together. The screen snap shot of LCRCvisual is one good example of the cross over of connecting links. Image you have hundreds of links running all over, it is very hard to check correct connections.
Screen size is the most prominent problem faced by visual programming language programmers. Their view are limited to what they can see in one screen full at a time. If the program is too big, the screen might not allow the programmer to view everything together in one screen. Therefore, more time has to be spent to ensure that they have made the right connections between building blocks. Recall that this is also one of the problems the future enhancement of KidSim is going work on.