I've found that if I set the "Text" semantic on one of the xhtml files, that Kindlegen will correctly translate that into mobi's "StartReading" value (EXTH type 116) upon conversion. I don't remember if that's enough for calibre to set the correct StartReading value when it converts ePub to mobi. But, I'm in the habit of using calibre's ePub Tweak utility to make sure the "start" item is in the reference section of the OPF. It's a bit of a PITA, but quick and easy once you get into the practice—and it will save you from having to rename your ePub to a zipfile and back and whatnot.
EDIT: Strike that about the "Text" semantic. It may work but I can't swear to it. I use a convoluted process that starts with Sigil... then uses calibre, mobiunpack.py, some manual tweaking, and finally wraps things up with Kindlegen. It works for me, but I can't swear that there's an easy semantic to set in Sigil that will cause calibre or Kindlegen to correctly set the StartReading value (without diddling it manually).
Last edited by DiapDealer; 06-15-2011 at 01:37 PM.
|