|
I'll apologize if my comments did not seem particularly clear, though they did not seem at all off topic to me, though maybe my understanding of the content server function is poorer than I realized.
As for my addendum in general I am surprised that I seem to be misunderstood, but I will make an attempt to break down some key issues.
Based not only on my own use, but also numerous complaints I have seen from others, Calibre seems to have problems with all of the following:
Loading/Organizing large numbers of materials ( admittedly a hard problem )
Inadvertent changes to the library ( even by automatic system processes )
File access ( especially where networks are concerned )
Storage ( primarily a file system issue )
As far as I have found, the only suggested solution to these issues has been to store everything locally, use multiple Libraries and to use the content server to access materials, neither of which address these issues at all. Feel free to correct me if any of this is untrue.
One suggestion made, perhaps unclear in purpose, was to allow parsing of containers other than folders as a method of reducing some of these issues. The reasoning is fairly simple, in that file system changes will not affect the internal structure of the container, and that performance might actually improve with the significantly reduced overhead of reading a manifest of a single file rather than individually accessing a minimum of 2 folders and 3 files per book.
Efficient alternatives to this somewhat more extreme than parsing a file as a storage location, if the Gutenberg Project comparison is any indication.
Edit: I will again say that I am not indicating a change in the structure of the Library, this point seems to be ignored too readily.
|