GNOME, please stop.

Posted Dec 27, 2014 20:35 UTC (Sat) by raven667 (subscriber, #5198)
In reply to: GNOME, please stop. by paulj
Parent article: Fedora 21 released

That may be in jest but there is a real grain of truth to this, there may be sufficient difference between ancient UNIX desktops and modern GUIs that to have sensible software you have to choose one style or the other and there are far more people familiar with modern UI conventions that ancient UNIX desktop conventions so there is a fundamental tension between supporting old UNIX hands and new people who don't have a UNIX background.

GNOME, please stop.

Posted Dec 27, 2014 23:06 UTC (Sat) by paulj (subscriber, #341) [Link] (2 responses)

I don't know if there's truth to it or not. That said, that Microsoft had to roll back somewhat from radical UI changes made in Win8 and re-introduce some long-standing UI metaphors suggests that perhaps there isn't too much truth in it. It may well be that the truth is that the desktop has matured, that people have gotten used to certain desktop conventions, and hence that there is little to be gained from radical experiments and much to be risked.

E.g., at the risk of degenerating into argumentum ad vehiculum, in the early days of cars, there were a variety of standards for the controls. Steering was not always via a wheel, some early cars had tillers; the throttle and main brakes were not always controlled by a foot pedal, but often hand-operated (e.g. a lever on the steering wheel for the throttle, a set of levers on the /outside/ of the car for the brakes; etc). It took time before cars settled on the UI conventions that dominate today -steering wheel, gear or drive selector in the middle, foot pedals for (clutch,)? brake and throttle actuation. Even light and indicator controls are somewhat standardised these days, perhaps thanks to consolidation in the industry. There is little value for manufacturers to experiment radically with this UI - it is mature.

I get the feeling though the GNOME devs believed in some truth shared with that comment. That they believed there is still value in big experiments with the desktop, or else that the desktop is dead and that GNOME must change to another UI (but, AFAIK, GNOME doesn't work well on anything but the traditional keyboard + mouse + decent screen desktop).

I'm not sure I agree with those things. I still appreciate the work the GNOME people do, but I've had to change the primary bit of my UI to Cinnamon in order to dampen how much radical experimentation I'm subjected to.

GNOME, please stop.

Posted Dec 28, 2014 4:45 UTC (Sun) by raven667 (subscriber, #5198) [Link] (1 responses)

To abuse your car analogy further the traditional UNIX desktop like twm or fvwm is like the old car with the external brakes and tiller steering, which some people are fanatical about, while the GNOME3 desktop is trying to be a next generation self-driver and there is a lot of conflict between the autopilot and tiller people.

GNOME, please stop.

Posted Dec 28, 2014 10:35 UTC (Sun) by paulj (subscriber, #341) [Link]

For me it's more about GNOME2 v GNOME3. I think most people were happy with GNOME2, least I was. The GNOME3 UI isn't more advanced, for me, it's just arbitrarily different.

GNOME3 isn't a next-gen self-driving car, for me, it's a normal car with the same controls, just in new, weird places. Some controls were hidden and are hard to discover - even commonly used ones like "open the menu" or "choose between shutting down and suspending". It took years for the GNOME devs to rollback on that latter one, despite the complaints.

Arbitrarily, unnecessarily different just isn't what I'm looking for these days in my desktop.

(And no, that some complained about the GNOME1 -> GNOME2 changes is not the same. The GNOME2 changes arose from empirical, objective HCI testing from which a GNOME2 HIG was formulated. There was clear evidence to say the GNOME2 changes were better for the majority of people. There was no such evidence based process to back the GNOME3 changes - it was a sort of experiment carried out on the user-base.)