Hi,

On 22. juni 2013, at 15:01, Timur Kristóf <[email protected]> wrote:
> About a week ago at Devaamo I was fortunate to meet with a bunch of Nemo 
> guys. Together we discussed and debated ideas to move forward from the 
> current situation to one where Nemo looks nice and is a pleasure to use.

My $0.02 on our current situation: in the area of home UX, we have something 
which is mostly functional, but lacking polish in some rather annoying but 
fundamental areas, like "I need internet access but there is none, so please 
show a dialog asking the user to set some up".

The main thing we've been lacking in that area is actually someone with 
graphical design talent. We need icons in a lot of different areas. Apart from 
that, it doesn't actually look too bad, and when I compare it to where we've 
been in the past, we're in very good shape now. That isn't to say that I think 
things couldn't be a lot better, but compare where we are *now* to, for 
instance, https://www.youtube.com/watch?v=xl7Djukoysc.

Application-wise, we're worse off. We just don't have the developers to get 
everything done at once. We're improving, gradually, but there's a long way to 
go.

> We feel that currently, the whole Nemo UI looks like just a cheap copy.

That's pretty much because it is. We inherited our roots and most of our base 
software from MeeGo, and a large cluster of users/developers from the same 
place. Most of our work has always been designed around being similar, with 
improvements where possible (and developer time has been available). An example 
of this is the landscape-mode calculator with history etc.

> We think we could do better than that.

I imagine I'm not saying anything new here, but keep in mind that designing 
something takes an awful lot of effort -- even with a team of 2-3 design guys 
working more or less full-time, it's still a lot of effort to have to crank out:

- platform design (the ideas/motives behind the design, how it ticks, user 
interaction design)
- application design (what they contain, how they work)
- graphical design (the production of actual assets used)

All of these, I'd say, are necessary if you really do want to do better. I'd 
really like to see someone try.

> Michael proposed the idea of making a whole new set of UI components for 
> Nemo, instead of continuing to use the deprecated MeeGo components, and we 
> think this is generally a good idea, especially now that we have Qt Quick 
> Controls in Qt 5.1.

Just a quick note: there's nothing deprecated about them, other than your 
calling them that. I'm still actively maintaining them, and will continue to do 
so for at least the foreseeable future. If there are bugs, they will get fixed. 
If there are considerable pain points, I'll do my best to address them.

Unrelatedly, as we should have learned from Nokia by now, just giving a finger 
to your developers and saying "learn this new thing" is not a very nice thing 
to do. These components were the developer story we had (and inherited from the 
N9), and we should continue to keep them working to keep existing applications, 
and developers, happy.

> With Qt Quick Controls, one can just use a set of standard components while 
> still being able to maintain a custom look and feel. This means that we can 
> create our own components without further fragmenting the already fragmented 
> world of Qt Quick APIs.

Despite the above, I would be very interested in seeing some effort taken into 
making controls usable over the longer term, as they gain us more 'free' effort 
and applications. I'd be a bit wary about jumping onboard straight away, given 
that they've only just seen an initial release, and given that our Qt 5-waters 
in Nemo are currently a bit under-tested, but nothing stops anyone from 
experimenting. Go wild!

> The other important thing is to think about the home screen experience.
> We think that the way forward in this regard is going for Wayland.

It is. :-)

> In other news, Robin and the others were able to get Wayland working on the 
> Nokia N9(50) devices

That's very experimental at this point, but yes, I'd like it to be usable one 
day. Hopefully some other determined people can start to poke at it, also… :)

BR,
Robin

Reply via email to