EVE Information Portal

...

This UI has been live for less than a month, and you've had roughly two months of feedback, and you're already adding huge kludges to fix problems that shouldn't have been there to begin with. This whole mess about “primary” and “secondary” inventory windows you've introduced only results in one thing: more windows, more duplication, more clicking, less intuitiveness.

Yes, you added back some rclick items to open special bays. Good job. Too bad they don't open in any kind of remotely sane location. They all go into the “primary” window, which means the functionality is completely redundant. The reason to have them in a rclick menu is to gain quick access to the bays — what happens now is that they act as a mechanism to slow down your access to other stuff, because opening them that way means you reset a window that is (probably) already being used for something else. The same goes for the cargo hold button — it opens my cargo hold in the “primary” window, completely overwriting what I had set it up to show.

None of these buttons are even remotely aware of the fact that I already have windows set up to show these locations, and they will happily open new instances/reset an existing window rather than focus on the window already showing what I'm after.

The old UI never had these problems because every inventory was its own window. Opening the inventory in question opened the associated window. Since the two are now disassociated, you will never really be sure of what will open where unless you do everything very deliberately and entirely manually (going through the whole shift-click-and-place-into-stack rigmarole), which is far more cumbersome under the new system than under the old one.

This amount of kludginess at this stage of the development phase — when you're still adding lost basic functionality back in — only points to one thing: the underlying design concept is fundamentally flawed. This is the kind of mess you'd see after years and years of quick-patching surprising edge cases (cf. CrimeWatch), not something you should see right out the gate, and definitely not something you should see as a result of just trying to make the core functionality work. You haven't even begun to touch on the edge cases, and we're already seeing end-of-life feeping creaturism and function confusion.

I've tried to stay away from the “roll back” line of thinking, but if the state of the code is this bad, maybe it's an option you should start to explore… or if not that, then declare this code path dead, much like how you did with the “Old New NeoCom” (the one that got shelved because CoreUI was around the corner) and try again, this time with the first-line goal of simply being able to do what the old UI could do.

http://youtu.be/RblUvkziO4A @2:45