Apple and Microsoft Have Stopped Taking UX Risks, and the Linux Desktop Can't Afford to Wait

In the end, using any device is all about doing actions (the simplest being reading/viewing, but also editing etc.) on data. Now the biggest difference between desktop and mobile is that the first one is (was?) file oriented while the second one is software (application) oriented.

Edit a document, listen to a music, on desktop you would historically navigate to where the file is, then open it with the program of your choice. On mobile, you would almost never do that, you would first open the app and from it select the file you want to interact with.

This is changing a lot how you are organizing the way you work. On desktop, it is easy to put related files of different types in the same folder, for example you organize your birthday, you can have the list of the invited people in a excel file with their number, a subfolder with the songs you want to play, some pictures of the venue, a text files with notes and etc. It is not natural at all to organize things like that on mobile. Android and iOS are hiding the files as much as they can. (This is not the case on Ubuntu Touch for example). And it becomes the responsibility of the app to organize the files it can open. This has a big impact on what you expect your desktop manager to done or not. I would personally be very happy if I could organize applications how we can organize files, and simply reopen a work session where I left it (some PDF open, some terminals open, my IDE, my browser...) even weeks after.

To complexify the t […]

The meta problem with all this desktop UX design discussion is that it's not actually very problem driven.

Many of us solve our UI problems and we land into our own happy minima.

Many of the UX "solutions" are either invented problems or not my problems or worse, grinding some anti-me agenda.

Like many, a lot of my UX improvements have been to simplify the UI. Tiling, with just a few windows per virtual desktop in herbstluftwm: kitty, emacs, browser and some rare transient application (usually PDF or image viewer).

The idea in the screen shot horrifies me. Why would I buy an expensive high pixel monitor just to use only 10% of it to display a single window? Why would I want to waste CPU to animate the migration of a full window to a single button icon (another example given)?

These UI "innovations" are great for movie props but do they actually help in reality? Not that I see. The "stagnation" of UI to me is more a plateau. Anything truly new must be driven by new "I", no input/output channels between user and computer.

The one interesting idea I saw in the article was Lifestreams. I think storage and recall is something that needs improvement and new ideas because for me that is a problem. I produce and consume too much info while recalling it is hard and while also I often do not produce or consume info that I wish I had.

Again, it's the I/O that drives the progress.

We could do well with change on the desktop, but very little of the change I've seen during the past 10+ years is actually benefiting the user. Even on Linux, the changes happening are stuff like bugfixes and reducing jank; small incremental improvements that have already been done, but are just getting their final round of polish.

All the other changes I see are just shuffling around the UI, which doesn't really help anyone. And even newer features are either so marginal as to not matter, or can't be trusted to stick around. Why bother to learn a new UI, when all that does is get you back to square 1, and it might not even stick around? I'd be much more interested in tweaking my DE if I could be confident in knowing that those tweaks will persist and not cause other problems down the line, but I have no such assurance.

I don't blame the casuals for hating UI updates. I myself went to Linux to get a more cohesive and stable experience. If we're going to make changes, we need to not break workflows that already exist. New features and workflows need to either not be in the way for established users, or opt-in until they're established and known good. Doing anything else is user hostile, and by this time, given how much we know users don't like changing UI, dare I say, evil.

For me it was a major revelation when it was pointed out that, for someone who doesn't want his desktop changed, I did install all sorts of crazy things when I was starting with Linux (Compiz plugins, anyone?). And while Gnome 3 pushed me into Mate, it wasn't because it was different but rather because it was bad. So I'm excited about someone trying something new, particularly someone who realizes that the Desktop is not a tablet.

Here's an idea that I believe I read in HN as "the original purpose of the desktop": add a Desktop Manager that manages which icons are on your desktop and makes it easy to switch.

The desktop is kind of weird, in the sense that it's prime real estate but, at the same time, there's an expectation that you shouldn't put too much stuff there, and then it ends up cluttered anyway. I think it's worth trying a stronger separation between "Desktop" and any other regular folder, emphasizing that the Desktop is where you place links to the files you need right now, but with a mechanism to clear it completely (without losing data) and another to switch which icons you see right now, kind of like browser tab groups. That way, when you're working on a project you place the files you need on the desktop and when you're done you clean it up.

One of the main problems with UX in my opinion is that someone arrives, sees a problem in their own eyes, and try to make changes and impose their views on everyone. Because a lot of that is personal preference, taste, habits, ability, and so on. You can't have a one size fits all solution. If you try you end up discovering that there are many dimensions and that people are scattered around in the space. And either you try to remove dimensions, and yay suddenly you have a one-size-fits-all solution but only because you removed choice (Google is a big offender in that regard), or you realize that you have no right to impose your views, get more humble, and propose a new solution, expanding the choice of others (but that is rare).

Man, there have been so many cool ideas for the Linux desktop in the past, and somehow none of them worked out or were polished into a usable state. Some things I remember:

* I think at one point Canonical tried to have a cool personal data sync with Ubuntu One using CouchDB which would have synced all your data between devices directly, not using a cloud?

* KDE 4 tired to have everything be a plasmoid (widget)? You could drag contacts to your desktop? But now we mostly have panels and tray popups again

* Similar, things like Evolution data server or Akonadi or Telepathy which promised to implement the "backend" for services like email, calendar and chat once and then you could integrate custom UIs into many apps and places in the desktop with a synchronized view (chat with a contact in your email programm? write a calendar appointment from you chat app that your friend sent you?)

* Maemo on the N900 had one contact list which served both SMS, Skype and XMPP and starting a chat using any one of those would get you to the same app/interface, same for the call, all happening in a single interface.

That is really the direction of "UX" I'm interested in. Not so much about where to display what, but to make my computing device feel less like an accumulation of applications and data, but to have a more integrated experience, ideally even across devices. My laptop and phone barely talk to each other and the user account on each device is distinct, so sharing resources already fails […]

I am surprised that Windows doesn't have the mentioned middle-of-name ellipsis. OneCommander uses it, but not Explorer. BitCommander had even smarter trimming prioritizing numbers and words reducing to letter plus dot http://bit-commander.com/GIF/bit-commander-file-name-shorten...

"Focus follows mouse" is what I use to keep the file manager from popping up when click-dragging files.

The biggest problem with UX developers is they know better than you AND force you into their paradigm. They came on the scene and decided "Choice is Bad" coupled with "Our way or the highway".

I've decided to be frozen in time with MATE (basically gnome 2). I have problems with it, but I've realized I don't want to be hit with surprise! on upgrades.