So Intel released a new version of Moblin with a totally new GUI aiming to bring the best use of those small screen beast, Netbooks and MIDs. While I was looking at the screenshots I was pleased by the fact that clearly there was some major design work there, not the usual theming work you see on opensource projects. For the sack of being positive at least at the beginning: FAST boot and clean iconography are the first impressions any user will have with Moblin.

Looking at the videos, I really thought:
” Damn they have something quite nice going in that project!”

Indeed they have, but my happiness endured only until the thing was fully installed on the Dell mini netbook.
Disclaimer: It’s a beta and a lot will evolve so I will focus my comment on the design decisions.

Once more: Shallow GUIs are evil

I have been calling shallow GUIs those graphical assets placed on the fore front of an Operating system in order to provide the fake sense of a real, deeply thought user interface. Like HTC did on the first batch of TouchFlo phones, like Nokia did with the “Multimedia key” menu and so on. They are all evil. Why?

Mixed experience most probably will lead to mixed consistency: While they try to innovate and bring something new to the table, they offer a limited reach, thus creating a set of inconsistency across what the user start using and what he ended up using. Firing a browser will take you to a more or less fitting UI but going towards any other application you will end up in the all old UIs done for the regular desktop linux.

How to solve it? This is not an easy task, but the goal here is really to create a GUI that goes beyond the home and the launcher. It needs to integrate all applications under the same concepts. What is clear by now is that Intel did the kick start but didn’t set all rules in a good way to be expanded across the system. This will lead to a lot of fragmentation, with new, modified GUI based on moblin emerging from other vendors.

Mixed input methods, same toolkit

Moblin is clearly aiming to target both netbooks and touchscreen based (MIDs) at the same time. Unfortunately this leads to other problem: Both become crippled in their own way. Netbook owners want content space while touchscreen owners want finger-ready UI. Moblin delivers HUGE buttons and controls that are a drag for the netbook owners. I actually tried to “like” them, as they make it quite easy to hit and use but one of the most important aspects of any netbook is screen space.

How to solve it? For me, the hardest one: This is a business decision. When you try to tackle 2 different enemies with the same weapon you will face challenges that I can only remember of in Megaman game series: Each enemy had a weak spot, and for that weakness there was the right weapon. You could still defeat it with other weapons but the amount of skill needed was way bigger. So if you business decides to solve this in the first week of concept, stand up and shout. We are dealing with the same problem (different input methods though) and it has been really stressing to simply decrease the experience of one input method because of the other. In the end, both will suck.

Abusing of 3D won’t grant you WoW-ness

It seems that in the iphone era everyone is trying to pump as much as 3D effects as possible. What they forget is that, even though, gratuitous effects are ok most of the time, you should be careful not to let them drive the application instead of the real goal for that particular feature. The photo viewer is one of the worst use of transitions ever made, and looking at the video is just impossible to imagine why the same designer who planned that strategy described in the video allowed this photo viewer to come to life. look at 4:10 minutes in this video : ( http://www.youtube.com/watch?v=2Ft0uoe3FpI )

And the list goes on an on…

Accordion-style menu for applications on Netbooks? Virtual desktops with windows that barely can be distinguished? No thanks. People will surely say “it’s a beta!” But the point here is really not about the software and it’s speed or stability. Is really about the real proposed design that is there. Once the ground work is set there’s not a easy turning back and fixing it. It’s not just a bug fix that will make the transitions smoothly. We are talking about the foundations of the user interaction in that particular OS.

Conclusion:

---

As much as I would like to say “Yes! UX coming to linux in a real way!” Moblin simple doesn’t do it. This award will still be granted to Palm and it’s WebOS. Moblin is doomed to be fragmented. It looks way better (designer POV) than Ubuntu netbook remix (and it’s REALLY shallow UI), but still failing to really think the experience in a deeper level, showing us only a first level of design thinking that immediately drowns in inconsistency when you hit the applications. Media x Browser x Simple menus are simple hacks with effects instead of “let’s also design those with the Netbook user in mind”

Update:

---

Here’s Linpus already customized Moblin UI. “I think HP’s GUI is cool, let’s add some of those. Look it fades between the panels”. Oh god.
—-
Already tested with: Dell Mini 9, HP mini 1000 and the MSI Wind netbooks. Currently trying: Windows7 on the MSI.