I believe you are examining the foundation of Computer Science. I consider the "Base" level to be binary but really it is the Turing Machine. If you can program a machine in its base binary level you can reach its maximum potential. Unfortunately, that is very labor intensive. "c" as defined by K&R was and still is the logical step towards a generic language. Few, if any limitations are introduced by the language. When you add the libraries, macros and other "enhancements" much efficiency is lost for the sake of the human interface. My favorite language is Pascal. Intended as a teaching language it will do anything but it is not difficult to severely tax the hardware with recursive procedures and dynamic storage allocations. For a while p-code (pseudo code) was popular. The idea was that everyone would write in p-code and an interpreter would translate to the intended platform. The result was akin to a bureaucracy, big time overhead! Then comes the Higher level languages. They usually target a specific area of interest at the cost versatility. There seems to be a pretty good choice of software-hardware combinations for today's interests. Basic Stamp and Rabbit have been the best combination for many enterprises facing high labor costs. Delphi and Visual Basic have grown to the point that they are no longer straight forward Languages by trying to be the general case solution for all occasions. ADA was a great idea that folded under its own weight. I believe the "Universal" language you seek is possible but not yet practical. If I were going to construct it I would develop a compiler directive that looks something like Set_Language (C,Pascal,Cobol,Java,VB,Delphi,C++,FooTwo); I like things the way they are. John Ferrell W8CCW "Life is easier if you learn to plow around the stumps" http://DixieNC.US ----- Original Message ----- From: "Thilo Klein" To: "Microcontroller discussion list - Public." Sent: Monday, July 16, 2007 8:06 PM Subject: [PIC] Recently > Recently I was talking about the problem of identifiying code within a > given project. The problem here is the same as creating additional > dimensions to a 1-dimensional system. It is not only the code offering > problems it is the sourrounding components aswell. > > I am glad to not be involved in development of a mobile phone or - > worse - a modern car. I think modern cars are mechatronic nightmares > anyway when it comes to error searching. > > Do you have an idea of standardizing software and hardware > combinations to make sure that almost any software is working on a > special surrounding ? Such as: "This software is running under > " becomes "This software is running under > ". And there should be only a few d. systems > > Some voices will arise and say: But this is the great thing about > mechatronics, you have more freedom in creating, you are even going > into the real world. > > There is a fascination about this topic, of course. Most people will > start to dote for the great opportunities, I mainly see a lot of error > possibilities for each single line of code that controls an outer > device. > > So how can we standardize it to make it simpler and this discussion > list understandable for those who do not own each single > microcontroller or at least knowledge about it ? > > Thilo Klein > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist