Panorama X file corruption


#41

Yes, I do, and your estimate sounds about right.

intensive development

For whatever it’s worth, you and Steve Middleton combined account for about 30% of all Panorama crashes.

Panorama 6 did not report crashes, the programming language it used had no ability to do that. I was quite excited about the ability to get detailed crash reports with Objective-C. Unfortunately, the results have been disappointed. Though crash reports appear to contain a lot of detail, most of them aren’t helpful at diagnosing problems. It’s kind of like getting a detailed layout of the arrangement of the deck chairs on the Titanic and trying to deduce the size, shape and path of the iceberg. I spent a lot of effort building the system that collects, transmits, receives and analyzes these reports, and there have been some significant bugs that have been eliminated. But it’s almost never possible to track down a bug from a single report.

If you ever encounter a reliable way to cause a crash, one that I can repeat while running under the Xcode debugger, that’s always the most helpful clue to fixing a problem. That’s far more useful than a crash report, but unfortunately of course there is no reliable way to repeat.

When I get an automated offer to report a crash, does that report go to you or to Apple? My recollection is that it’s Apple.

It goes to Apple.


#42

Brilliant!!


#43

I suspect that I will be able to do this - there is some level of constancy about my crashes so I’ll try to be more aware of them. As you can appreciate, when I’m beavering away at a complicated task I’m more interested in not crashing!


#44

Im sure that I am probably the majority of the 30%. :slight_smile: I have been developing a large solution as well which has challenged my programming abilities. I have also tried almost all of the objects and data formats, ie arrays, data dictionaries, etc. while experimenting with different approaches to solve a problem. I have expanded my skills through much trial and error. I can confidently say that most, if not all, of my crashes were operator error and not Panorama X’s fault. Once I fixed my errors, crashes are mostly gone and the ones that do still happen are my errors that I have not tracked down yet or have not had time to fix.


#45

Ultimately, all crashes are Panorama X’s fault. Even if there is operator error, Panorama should respond with an error message or some sort of recovery, not by crashing.

I’m glad to hear though, that the crashes are “mostly gone”. Looking at the data, I can confirm that the rate of crashes you are encountering in 2019 is down by over 65%. Perhaps less trial and error at this point in your development.