MobileRead Forums - View Single Post - Would you benefit from file manager on your Kobo reader?

BWinmill · 06-19-2013, 05:14 PM

I can understand the desire for a file manager from a few different perspectives. On the other hand, I think there are better ways to do it.

A file manager is easily understood, at least among those of us who have been using them for years and know how to organize our files. File managers are also universally accessible on personal computers, so you can organize books using the same skills and strategies that you use to organize your other documents. There are also very few boundaries in how you structure your directories as well as how you name your files and directories. The main limitation is that it must be hierarchical, and even then you can create links to sort-of get around that limitation.

Yet I don't buy into the bit about it being easier. Books contain metadata. As a minimum, the author and title are virtually always correct. The cover and other publication data are usually valid as well (e.g. publisher, publication date, series information). Tags are usually dodgy, but let's ignore that for a moment. Let's pretend that your main concern is organizing books by title, author, series, and genre. Let's look at the difference between using a directory structure and using the metadata.

When you download a file, the directory structure is not there. You need to create it (a one-time job) and you need to populate it (an ongoing job). Filenames may give you the title, author, and series. It is unlikely to provide the genre. Yet both are likely mangled. If you obtain all of your books from the same source, you can probably get around that with a file manager that supports find/replace (as a minimum) or regular expressions (ideally). That file manager would probably place the files in the correct directories as well, and will do so with the minimum of intervention. If you obtain your books from multiple sources, the above holds true but you'll probably have to select which rule(s) apply to a particular source. Then there are the sources that provide virtually no information in the filename, such as Project Gutenberg, so no file manager is up to the task. Everything has to be done manually.

Using the metadata is altogether different. The drawback is that you need special tools. Calibre is the classic example, though there's no particular reason why you couldn't use a special file manager (or file manager plugin) that recognizes ePubs or a programming library that recognizes ePubs (e.g. a Python package). But all of that is too complicated, so let's pretend that you're using a dedicated tool like Calibre. This tool will instantly know about the author, title, and series data. The tags may include the genre, or you may have to add that yourself. Your work is effectively done because the work has been mostly done for you. You don't have to unmunge file names. You don't have to put things into directories. It is neither a manual task, nor one requiring the user to automate process in that aforementioned powerful file manager. At most you'll have to go in and add genre data. At worse, the picky person will have to make minor corrections to the data.

06-19-2013, 05:14 PM	#95
BWinmill Nameless Being	I can understand the desire for a file manager from a few different perspectives. On the other hand, I think there are better ways to do it. A file manager is easily understood, at least among those of us who have been using them for years and know how to organize our files. File managers are also universally accessible on personal computers, so you can organize books using the same skills and strategies that you use to organize your other documents. There are also very few boundaries in how you structure your directories as well as how you name your files and directories. The main limitation is that it must be hierarchical, and even then you can create links to sort-of get around that limitation. Yet I don't buy into the bit about it being easier. Books contain metadata. As a minimum, the author and title are virtually always correct. The cover and other publication data are usually valid as well (e.g. publisher, publication date, series information). Tags are usually dodgy, but let's ignore that for a moment. Let's pretend that your main concern is organizing books by title, author, series, and genre. Let's look at the difference between using a directory structure and using the metadata. When you download a file, the directory structure is not there. You need to create it (a one-time job) and you need to populate it (an ongoing job). Filenames may give you the title, author, and series. It is unlikely to provide the genre. Yet both are likely mangled. If you obtain all of your books from the same source, you can probably get around that with a file manager that supports find/replace (as a minimum) or regular expressions (ideally). That file manager would probably place the files in the correct directories as well, and will do so with the minimum of intervention. If you obtain your books from multiple sources, the above holds true but you'll probably have to select which rule(s) apply to a particular source. Then there are the sources that provide virtually no information in the filename, such as Project Gutenberg, so no file manager is up to the task. Everything has to be done manually. Using the metadata is altogether different. The drawback is that you need special tools. Calibre is the classic example, though there's no particular reason why you couldn't use a special file manager (or file manager plugin) that recognizes ePubs or a programming library that recognizes ePubs (e.g. a Python package). But all of that is too complicated, so let's pretend that you're using a dedicated tool like Calibre. This tool will instantly know about the author, title, and series data. The tags may include the genre, or you may have to add that yourself. Your work is effectively done because the work has been mostly done for you. You don't have to unmunge file names. You don't have to put things into directories. It is neither a manual task, nor one requiring the user to automate process in that aforementioned powerful file manager. At most you'll have to go in and add genre data. At worse, the picky person will have to make minor corrections to the data.