On 20 Oct 2005 at 5:56, Bernard Savoie wrote: > On Oct 19, 2005, at 13:00, [EMAIL PROTECTED] wrote: > > Hiro wrote: > > > > One more thing. Even though the bug of loosing data is pretty bad, > > you can't call it file corruption unless the file cannot be opened > > without error message. > > Sorry to disagree with this one, corrupt for me means that the file > does not deliver the material which was saved to it, whether you get > an error message or not.
>From a database point of view (and Finale files are database files), it's definitely corruption, and of the logical type. There are two types of corruption in a database: 1. logical corruption 2. content corruption. Logical corruption means that the data structures in which your information is stored are corrupted. It's quite clear from someone's post yesterday (can't remember exactly who it was) who did a file integrity check that the internal structures are completely hosed -- a whole lot of pointers connecting data to their usage information has been lost. Content corruption refers to a different kind of corruption, not structural to the storage format, but inherent to the information being stored in the data structure. If, for instance, you copied a passage of music over an existing passage, the music (the content) would be corrupted, though the Finale file is still structurally 100% *not* corrupted. It's the information being stored that is now out of whack with what's intended, and that's a different type of corruption. > So until this problem is diagnosed, I would call a file having this > occurrence, perhaps not corrupt, but potentially corrupt and > certainly suspect. There's no question that it's a form of corruption of Finale's internal data structures and it's causing loss of data. -- David W. Fenton http://www.bway.net/~dfenton David Fenton Associates http://www.bway.net/~dfassoc _______________________________________________ Finale mailing list [email protected] http://lists.shsu.edu/mailman/listinfo/finale
