I spent some time with Ardour, a few years back; Paul Davis' ProTools semi-clone for Linux.
I did a project with it, and discovered that it's UI didn't follow any of what we consider to be the standard rules that such interfaces should follow for usability, discoverability, and suchlike.
And, of course, I got on their IRC channel (or mailing list; I forget which it was), and embarassingly, ranted a bit more than was perhaps really called for.
To the everlasting credit of Paul and a couple of his lead developers, they were very polite to me, and over the course of a few hours of editing, they led me to see that the ways in which they'd broken the customary interface design guidelines we've all come to know and expect were done for very good reasons of efficiency and ease of actual use by professionals doing professional work -- which is the target of Ardour, if not of some other programs like Audacity (which are very nice, but are not targeted at Bob Clearmountain and Nile Rogers).
I don't know whether this potential explanation for the way Darktable is designed might be serviceable -- I've never used any of the commercial packages either, much less DT itself, so I can't speak to whether it's trying to be 1-2-3 compatible with one or not.
But I've learned (!) not to assume that something's poorly designed *solely* because it doesn't follow The Rules As I Understand Them.
Posted Jan 23, 2012 19:02 UTC (Mon) by n8willis (editor, #43041)
[Link]
Well, I appreciate your comment, but I did not "assume" anything about Darktable's UI or abstract rules-as-I-like-them. The unlabeled icons and the infexibility of the adjustment widgets I mentioned are demonstrably detrimental to the usage of the program in practice. You *could* memorize the names of all of the unlabeled icons, for example, but the fact that you could do so does not mean that the icons successfully communicate what they do.