How to screw a release

Two days before SimpleMovieX 3.8 was released, I was approached by a potential customer, Jon, that said he would buy my app if I added a small feature: drag multiple clips at the same time and stitch them together in order.
SimpleMovieX could already do that, but you had to drag them one by one, and paying a lot of attention to where the playhead was or the movie would be inserted at the wrong place.

Needless to say, I had already be warned against those “last minute”, “sneak-in”, “one line of code” features. And I had also experienced it the hard way a couple of times. But the flesh is weak, and I finally included it. It worked great.

So great that Jon saw a possible improvement: order the files by name or by dates. A few more line of code, and it was done. Even greater, if I could add chapter markers automatically to make the multiple movies visible once they’ve been merged. Yes, Jon I can do that. It’s just one more line of code.

And finally, I proudly released version 3.8, advertising “Merge several movies in one shot” as one of the top improvements. There’s just a small problem. This “one more line of code” that creates the chapters on the fly, also breaks a much more important feature: Native saving, the capability of saving the merged movie in avi or mp4, whichever format the original movies had.

Don’t get me wrong: I want more Jons, more customers with great ideas to make SimpleMovieX better. Thank you Jon.
But the bottom line is: Many new customers will be turned down when they try the “merge in one shot” and discover that they can’t save. What failed? Validation, due to overconfidence. Never underestimate one line of code. Even in small apps like SimpleMovieX that you’ve written end to end.

PS: I’m currently working on the fix, it’s almost done. There’s a simple workaround: Drop your files one by one as in previous versions. Once I’ve a fix, I’ll publish a download link.