My initial reaction to the beta version was that the UI was much worse, but I decided to give it a chance and spend a few weeks trying to get used to it.
Now that I've had time to play with it, I'm even more convinced my initial reaction was correct. But I've also had time to reflect on why, and in my view there are two problems with this new design.
Removing visual separators from the tabs makes them much harder to distinguish - the only difference between one tab and the next is a larger horizontal space. In prose, when we want to separate sentences, we don't just use longer spaces; we use other visual indicators like capital letters and full stops (periods). Paragraphs are separated by vertical space, which is visually distinct from a horizontal one.
The second problem is that the tab you're not supposed to click on (the active tab) looks like a button, while the tabs you are meant to click on (the inactive tabs) are presented as text. This is opposite to normal UX conventions.
There are some subjective improvements to the look and feel of this UX change, but overall I'm confused as to why the Firefox UX team would go against established conventions to produce something that's visually more difficult to use and less intuitive than they had before.
Fortunately userChrome.css is still usable if you set an about:config option (toolkit.legacyUserProfileCustomizations.stylesheets), so I can see about fixing the problem myself, but I'm concerned that Firefox's march toward reduced configurability will eventually remove this option.
Overall, my concerns on Firefox are the same as most; in the pursuit of market share, they're moving further and further away from their roots by removing customisation options. Remember when you could change Firefox's theme without using hidden configuration options?
