Delete files containing no named commands.

[bengtmartensson]

There are a number of commands with no names in the data base. Such a command is like a phone number which you do not know to who it goes. You may be able to do "further research" (like calling that number), but until then, it is a useless piece of information.

This PR deletes all files that contain no named commands.

[probonopd]

Well...

You may be able to do "further research" (like calling that number)

Isn't that reason enough to keep them?

[bengtmartensson]

No, it is not. Work in progress should be clearly recognizable as such, and not just kept together with deployable data. For example, moved to a separate branch, either in the file system or in Git. This is just normal good manners in SW engineering, see for example "Uncle Bob" (Robert C. Martin: Clean code).

The particular nature of this very data makes "further research" extra problematic. Most of the questionable files were submitted by people who in most cases are no longer around, and the devices are also probably recycled a long time ago.
Actually, if someone wants to repair some of these files, it may be faster to start from scratch.

The fact that the project does not keep any more meta data than contained in file names is another problem.

If you (or someone else) are willing and able to repair some of the files I suggested for deletion, fine.

[bengtmartensson]

Simon, how are you going to resolve this? Keeping knowingly useless stuff (at least for the user) in a data base intended for the public is not a good idea. Alternatives would be to shove it to a (Git) branch, "unfinished" or so. Or you can create directory, say "_unfinished" and move it there.