Lost Text Document on iPad when Battery Drained

I’m a long-time DT user, and generally very happy with the software!

Today though… different story (on my iPad Pro w/iPadOS 17.2 running DT 3.8 (17416).

I created a new text document, and spent over an hour entering experimental test results into the file. I left the document open the whole time. Occasionally DT popped up a dialog saying “You have unsaved data. How would you like to proceed?” The choices were Keep and Delete, and I always tapped Keep.

Unfortunately I let the battery run down, and the iPad shut off. I plugged it into a charger and re-launched DT, but the text file was nowhere to be found, even though I had been clicking “Keep” repeatedly!

I searched pretty thoroughly, and did not turn up any sign of the file. So essentially all my experimental data is lost!

If anyone knows of a hidden location where I might find the file, that’s great. But mostly I wanted to report this as a likely bug, for the developers to consider some “fix”.



This technically is not a bug. The New Document Assistant (NDA) isn’t intended to be left open for editing over extended periods of time. It is a tool for beginning and saving a document so you can continue using the other features of DEVONthink To Go. This is also supported by the fact you can’t recall the NDA to continue editing an existing document.

Pressing Keep is not saving the document. It is preserving the existing content in the NDA instead of purging it to do something different in the application.

all my experimental data is lost!

While it doesn’t solve your problem, at least it wasn’t a loss of important data.

Hi Jim,
Thanks for the explanation. With this awareness I can avoid the issue in the future.

But for other users who aren’t aware of this “feature” of the NDA, it would be helpful if that dialog in DTTG had a big red banner to indicate what action must be taken to save the document, and let the user know that “Keep” doesn’t mean “Save”. :slight_smile: I suspect I’m not the first person who lost data using the NDA.

Regarding the experimental data, it was indeed important (to me). I was conducting experiments with the intent of analyzing the data later.



You’re welcome.
Development will surely see this thread and have things to consider for improvements to the NDA.