[SOLVED] Bah! Years of work wasted!

[hazel]

No, it wasn't because I didn't back up my data . This is something much more fundamental.

Some of you will know that for years I've been tinkering away at what is intended to be a music storage and writing system. At the centre of it is a file of structures representing notes and other musical glyphs, which is created through a visual interface - pinning notes on a stave on the screen. The file can be printed out as sheet music, and I was just beginning to work on a program for editing the data. Later I intended to write programs for playing the music (for auditory "proof-reading") and for transposing it into other keys.

Last week I updated Debian Wheezy as usual, and then looked for whatever was new in the repository. I found a program called Philip's Music Writer and installed it. And it does everything I was planning to do only much better! Instead of an opaque file of binary structures, it uses a text file that you can create in an editor. PMW turns it into a postscript file of music which you can print. You can also transpose it and you can play it through midi.

Just for interest I include a text file and the pdf that PMW makes from it.

I was using my system to teach myself C programming but now it seems pointless to continue. Grrr!

[elija]

I was using my system to teach myself C programming

I feel your pain, but that doesn't seem pointless to me

[sgosnell]

Not a complete waste, I would think. At least you now know something about C, and have more ideas about how to implement a system, so you might in the end be better off than you would have been without finding Philip's work.

[Lakshmipathi]

My suggestion would be complete your work regardless of anything else. Someone may find your coding style is more easier to follow ,thus it would allow them to writing a new one or modify/enhance your work.

Recently when i was search for "ext" fs code found that there was another file system named "xia" was also there. Both ext and xia competed for default linux file system.eventually ext has better feature than xia , so people opted it. But still I would like to check the source code of xia to understand basics of FS.
Xiafs - Wikipedia, the free encyclopedia

[hazel]

I've just realised something. If I can get the editing bit of stavescribe to work, it could easily be rewritten to act as a graphical front-end to pmw. All I would have to do is make it create and write pmw codes instead of glyph structures. More sophisticated editing could be done with vim or gedit but the basic file could be created that way, and I think that would actually be more intuitive than writing all the codes by hand. And it would be a sort of revenge on Philip for pipping me to the post.

[kohog]

Before you waste more time, you should take a look at these

Lilypond
MuseScore
NoteEdit
Denemo
Rosegarden
Frescobaldi
LilypondTool

[Bemk]

Dear Hazel,

I say, that if it is fun doing you should continue your work and nothing but the fun in important. If it isn't any fun, don't do it and use the alternatives.

I do it all the time! (Why else would I be writing my own OS?)

[Rubberman]

Reinventing the wheel is not a waste, if you end up with a better wheel! I like to work on garbage collecting heap memory allocators. There has been plenty of work on these by others, but I use the work as a means to better understand how allocators work, and how they can be made more efficient and secure. I wrote a reference-counting deterministic time behavior GC for C++ a number of years ago. It allowed a 10M LOC manufacturing execution system to be written without any deletes, no memory leaks, and none of the stuttering behavior you would see in Java with its standard mark-and-sweep collector. I still noodle on one for C so applications can be made safer from bad pointer dereferencing, eliminate memory leaks, and still perform in linear time.

[hazel]

Well, yesterday I fixed another little bit of the system that involved setting, testing and unsetting some alarm flag bits. I never could understand bitwise arithmetic before and now I do. So I think it's worth continuing simply as an educational project.

[Rubberman]

Well, yesterday I fixed another little bit of the system that involved setting, testing and unsetting some alarm flag bits. I never could understand bitwise arithmetic before and now I do. So I think it's worth continuing simply as an educational project.

I always thought of bit-twiddling as a bit of a dance, sort of like the cha-cha. Shift left, xor with your partner, shift right, mask the fiddly bits, and away we go!