@gksloane wrote:
To actually document the fact that instead of coming up with a reasonable way to identify failures caused by lack of runtime support at runtime, you state that you might return an error, or you might just run code incorrectly, isn't a reasonable way to design and distribute a commercial product to customers.
Emphasis mine.
Please be aware of the fact that this is primarily a user-to-user forum. No one responding to you thus far works for NI. (One would hope they would be a bit more polite if they did.)
@gksloane wrote:
Just complaining.
“Unknown Error” means either (1) the error was handed to me from somewhere else, I didn’t generate it, and I don’t know what it means; or (2) I’m too lazy to write an error handler that returns meaningful errors.
The language I'm complaining about is "LabVIEW returns an error or unexpected results...". There's a big difference between returning "Unknown Error" and returning no error with unexpected results... If I were to put that language in the software specification on any project I'd get laughed out of the room. So instead of documenting it, NI should fix it.
BTW, the reason I even bothered to post this is because NI solicited feedback on their new document center and document format; this happened to be the first document I clicked on. No good deed goes unpunished.
Did NI state that the LabVIEW community discussion forum was how to deliver that feedback? You'll be somewhat lucky if anyone from NI really notices this. If you want to have slightly better luck, or at least give other users the opportunity use a type of voting mechanism on your suggestions, try posting something in the Idea Exchange. If you prefer to speak directly with NI, you should open a support request.
Edit to say that I think I just got the same popup you are talking about, which said to leave feedback on the documentation center here... Although I suppose this isn't actually feedback on the doc center anyway. 🤷
For what it's worth, I don't think you're wrong. "Unexpected results" is certainly different from an undefined error, and slightly concerning.
That said, using VI server in that way is a somewhat advanced topic. MOST of the properties and methods not supported in the RTE are ones that don't really have many, if any, good use cases in a built application, and most of them are also not visible without specifically turning on the option to see them. (Scripting also wasn't meant to be available to the general public at one point.) Not that all that completely absolves NI, but I would generally expect users capable of competently using those methods to also have a good grasp of their use cases and their limitations.
