[EpiData-list] EpiData Notes file

epidata-list at lists.umanitoba.ca epidata-list at lists.umanitoba.ca
Mon Aug 16 11:37:54 CDT 2010


In my quick test, the largest note file I could create was a bit over 320,000 bytes. I could read this back with no problem. But Enter just would not let me add to the file. I doubt that your file was that big from only 40 subjects. I've never used .not files for qualitative data, but this is an interesting idea that gets around the limitation of string fields. .not files don't offer any functions for organizing this data, however.

If I'm right about your file size, it sounds like something else corrupted your file - for example, by introducing non-printable characters or through a disk error. I haven't used .not files enough to ever see any problem with them, but I've had lots of files corrupted by Windows over the years. When a file seems to disappear right in the middle, it is usually a disk error or other file corruption.

***The safest recovery is from a backup - either a routine system file backup or by creating your own EpiData backups. Backups can even be automated to some extent with EPIC, for example.***

In a MAC environment, TIme Machine backs up everything I do. My experience with system backups in a Windows or networked environment is not a happy one. We lost, randomly, about a quarter of our old LAN files when a corrupted system backup was the new starting point after a catastrophe in the server room. Your best bet is to save your own removable disks or have a backup folder for your project.

In principle, your hard drive may still contain a copy of the missing data in space that is not being used, but these days, it is probably quite difficult to recover yourself. 20 years ago it was quite straightforward.

Jamie Hockin

On 2010-08-16, Suzanna wrote:

> We set up an EpiData database with a notes file <filename>.not. The
> notes file appears to have corrupted and has taken the qualitative data
> from over 40 surveys with it.  I was wondering if there was any way I
> could recover this or if I'll need to re-enter all the data manually
> into a word document.
> Do you know of some recovery technique for this kind of problem?

More information about the EpiData-list mailing list