Here's an interesting post on the SourceTree development blog about why they believe their changes to the UI in the last few years are better, and here's another than explains changes to the Windows client specifically.
I get the sentiment of some of the changes, like this one...
The Windows App always used tabs for navigating between open repositories, but we knew that those tabs weren’t quite right. The toolbar sat above everything else in the UI, but it only really belonged to the open repo (tab). We’ve swapped those panels so that tabs now live at the top of the hierarchy, much like browser experiences that you are no doubt familiar with.
... but I don't necessarily get the conclusions. You want to have a coherent design philosophy, like Material Design or Apple's User Interface Guidelines, sure. But I've never noticed tab placement to be a problem in SourceTree. I get that it's not a browser look, but also wonder why we need to standardize on browser behavior. Is overuse of tabs in Windows apps a convention in itself? Probably was. Do newer Windows users balk at that convention and prefer browser conventions? Maybe. I don't know. I didn't see a lot of A/B testing in the SourceTree articles.
What I do know is that I hate the flatness of the newer SourceTree. I also hate how spaced out each line for a checkin is. It's disorienting and not as info rich as the alternating colors version we had previously. I also dislike the icons in the toolbar. Before, the icons were colorful, but not distracting. Now they're as flat and non-descript as a Jony Ives' designed button on iOS 7.
I posted a discussion on their forums, not that I expect to see much in the way of responses.
Just for fun, I thought I'd ask if anyone else preferred the UI from back in 1.7. To my eyes, at least, it's much less fatiguing to take in at a glance than the flatness of more recent releases, with nice, though old-school, cues about where widgets start & stop. Sort of like having serifs in a font.
The colors are also more muted, giving information rather than everything demanding attention at once in an overly egalitarian fleet of brightness floating on a flat sea of white.
(I do kinda like that last line in the quote, above.)
But one of the neat things about their own blog posts is that they do seem to care what you think. In that first that I linked to, you could schedule a 1:1 with them to discuss what you thought about design. Let's just go ahead and say that's crazy customer attention. ;^) And they ask for community posts if you have an opinion, so I'm doing my part, I guess.
Makes you wonder, though.
- How much attention and how much of a company's resources are put into UI where it's not useful?
- How can you tell?
- Do you embed metrics in your apps once they're released to measure real-world use?
- Is real-world A/B testing enough?
I mean, Linux still hasn't taken over the desktop for vanilla PC users. Why is that? Just apps? I'm betting there's some UI roughness that's also a barrier to entry. Good enough [for functional use] isn't always good enough [to sell (or be eaten?) like hotcakes]. And it's specifically these kinds of changes... less information, buttons being moved to different spots... that are the microactions that make everyone's day to day that much less productive.