🚀 go-pugleaf

I've been reading the code in the Interviews 3.0 distribution for about
a week now, and I'm distressed at the lack of error checking code,
specifically in the Doc application.

It seems to assume that every call to every function will succeed, and that
all input files are in the correct format.

Granted, this app is advertised as a "proof of concept," but I'm afraid
I can't use any of the classes in the doc application because they are
so optimistic (and we know Murphy's rule reigns supreme in the computer.)

I'm going to look more closely at the interviews code to see how it handles
errors.

Also, all the X-dependent code is supposed to be down in the IV-X11 library,
but I see lots of X-dependent code in apps that use fonts. They all use
R4 LFD font names. In general, fonts are not handled well.

Whose brain-dead idea was it to limit fonts to 75dpi fonts? There's code to
compute the real size of a pixel, why not use the same technique to find the
real size of fonts?

I was hoping to use Interviews in production code, but I'm worried about
the robustness of the code. Somebody please tell me it's just the sample
apps and not the libraries that have this blatant disregard for errors.

Dan