| And most of them aren't. And even those that are - it's much easier to implement the ability to retrieve the last chunk of file than to deal with significant performance degradation of forced file rewrites. Think about a format that has all those properties and you've used - PDF. PDFs the size of several 100s of MB aren't rare. Now imagine how it works in your world: * Add a note? Wait for the file to be completely rewritten and burn 100s of MB of your data to sync to iCloud/Drive. * Fill a form? Same. * Add an annotation with your Apple Pencil? Yup, same. Now look at how it works right now: - Add a text? Fill a form? Add a drawing? A few KB of data is appended and uploaded. * Sign the document to confirm authenticy? You got it, a few KB of data at the end. * Determine which data was added after the document was signed and sign it with another cert? A few bytes. Do you need to stream the PDF? Load the last chunk to detect the dictionary. If you don't want to do that, configure PDF writer to output the dictionary at the start and you still end up with a better solution. |