Hmmm....funny, they just had a big discussion on the SuperCollider list about whether or not more "Windows-like" sanity checks to save the application from programmers who don't know how to program was necessary. Usually such "fool-proofing" kinds of checks come at an added expense in terms of CPU overhead. Some people went on the side of "people should read the manual and know what they are doing", and some thought the application should work harder to save them from themselves, and they were ready to accept a certain decrease in efficiency to do so.

It is worthwhile not to forget that PD is a programming language, and that it is used to write applications. So erroneous input from the programmer is perfectly capable of crashing without recovery...i.e. patch author=programmer, not end-user. In my own little chunk of reality, if you author a patch that is tested for all versions of PD and all OSes, then it is your responsibility to make sure it is robust enough to withstand "stupid user tricks". But that's what makes PD so damed fun, innit?

my two-euro-bits, d.

Matteo Sisti Sette wrote:

F. Barknecht wrote

...

Well, that's an error of the patch-author.

Well, whenever a program crashes and can't recover (that is, it forces the operating system to close it), it is ALWAYS a bug of the program (unless it is of the operating system of course) no matter how erroneous the input from the user (i.e. the patch author).