Well, I pretty much know but… it's so much work, and some open-source and multi-platform software is available; TuxGuitar is pretty good, for example (
http://en.wikipedia.org/wiki/TuxGuitar[
^],
http://www.tuxguitar.com.ar/[
^]). So, I don't know why would you do it if you have not enough experience.
You could think about some software development is you had absolutely original or innovative ideas. I believe there is an unlimited field for inventions, but repeating GuitarPro… why?
I have developed some experimental stuff, but it was really interesting to me. Besides, I think the role of such tools is somewhat exaggerated. Well, they are good if you want to exchange compositions or to digitize opuses previously available only on paper. Many people try to learn music using such tools in hope to make it more effective, but in practice it does not pay off so well. You see, you waste so much time on input of notation, the time you could spend more effectively. Take learning playing guitar, which is quite a difficult thing, if you want to produce a little better impression then a march cat. :-)
At first, you need to do some seemingly boring calculations, so a pen and paper help a lot. But as you do it, you learn musical relationships between sequences and chords, essential, not mechanically remembered chord structures, and trying to find application of a guitar neck, you learn the note layout on the neck. And you learn to pick melodies and harmony by year, can pay attention to the manner of phrasing. If you use something which shows you something like graphical tabs with position of fingers, you miss all that; it provokes you to fall into mechanical remembering and mechanical reproduction of musical motion. Many people learned how to reproduce known sequences to several songs and got stuck there. Still, stuff like GuitarPro or TuxGuitar could be useful, but not often.
—SA