So, I noticed that this page talks about the need for an organized system for registered filetypes. I thought of a good one, and I want to ask if this'll be feasible (after, of course, we leave the existing filetypes as they are).
When we create our program filetype, we get three bytes. How about the first two bytes are simply the creator's ticalc.org author ID? Thus, for instance, Kerm would always begin the filetypes for his programs with 6077, and I would always begin my own filetypes with 10781. Then, the last byte is whatever the programmer decides it should be, so each programmer is alloted up to 255 unique filetypes. Because the most recent author on ticalc.org only has an ID of 11210, we're still decades away from 65,535. And the 255 filetypes shouldn't be a problem because no one has even published over 255 files in total (except for Kerm and the author of Phoenix, and a good chunk of those aren't 83/84 Plus).
The already existing prefixes would be an exception to this rule, and would instead be occupied by universal standards for files, such as plaintext or a 768-byte image. That way, no one programmer would have to declare the filetype themselves.
Please tell me what you think! Should we use this?
When we create our program filetype, we get three bytes. How about the first two bytes are simply the creator's ticalc.org author ID? Thus, for instance, Kerm would always begin the filetypes for his programs with 6077, and I would always begin my own filetypes with 10781. Then, the last byte is whatever the programmer decides it should be, so each programmer is alloted up to 255 unique filetypes. Because the most recent author on ticalc.org only has an ID of 11210, we're still decades away from 65,535. And the 255 filetypes shouldn't be a problem because no one has even published over 255 files in total (except for Kerm and the author of Phoenix, and a good chunk of those aren't 83/84 Plus).
The already existing prefixes would be an exception to this rule, and would instead be occupied by universal standards for files, such as plaintext or a 768-byte image. That way, no one programmer would have to declare the filetype themselves.
Please tell me what you think! Should we use this?