The core of BrailleBlaster IS in C. It is liblouis and liblouisutdml. The latter will eventually be a complete braille transcription engine, capable of doing whatever a place like APH or NBP wants. It is already being used for production in Europe. It was written as a library so it could be used in any nuumber of applications. In my view, BrailleBlaster is just one of those applications. I like to say that liblouis-liblouisutdml is the engine and BrailleBlaster is the rest of the car. You can put the same engine in many different makes of car, but an engine doesn't go anywhere by itself. BrailleBlaster IS the UI. Any high-level language would require a runtime environment. Alex is the only one who is talking about bundling JREs. I think most users would already have one for their machine. If not, the installer could fetch an appropriate one. The pure Java code is platform-independent and does not have to be recompiled. Only the native libraries must be compiled for the p0latform and the architecture. From what I've been hearing, computing is moving into the "cloud." That is why using the SWT browser control is of interest. BrailleBlaster was never intended to run on embedded devices. Even liblouisutdml would probably be too big for them. Some are already using liblouis. I don't think that there would be much difficulty with BrailleBlaster on Linux. There is a gtk versionn of SWT and probably others. Quite frankly, I've been working on the Java code because nobody else is. I want a good car to put my enginne into. I would rather concentrate on the engine. If somebody wants to put it into a different car that is fine. I just think that BrailleBlaster with Java is the best bet for building a driveable car. John -- John J. Boyer; President, Chief Software Developer Abilitiessoft, Inc. http://www.abilitiessoft.com Madison, Wisconsin USA Developing software for people with disabilities