Tidbits from the Usability Hackfest

« May 2015
Mon	Tue	Wed	Thu	Fri	Sat	Sun
				1	2	3
4	5	6	7	8	9	10
11	12	13	14	15	16	17
18	19	20	21	22	23	24
25	26	27	28	29	30	31

If you're still wondering what happened during the Usability hackfest, then you clearly missed a lot of blog posts. The good news is that you can catch up with all the links being collected on the Hackfests, or you can cheat and go read Máirín's coverage, since she did an amazing job writing about what was being discussed.

I was there only for the last two days; the original plan was to attend a bit more of the event, but the travel from France to London took an unexpected 12 hours. I still had some good and useful time there, that I mostly used to get a good overview of what people are working on, and how this can be integrated in a GNOME roadmap. Here are some highlights:

While I missed the discussion about nautilus, it seemed most people at the hackfest agreed on streamlining the nautilus user interface. I'd love to try the prototypes that were worked on: most of the proposed changes make sense to me. But getting rid of tabs and/or the split view will certainly trigger various reactions, and that's something that we cannot ignore...

Thomas je parle français couramment Wood was kind enough to let me use his laptop charger nearly all the time — I had one, but not for the right laptop...

Charline reported about a usability review of empathy, and this was definitely instructive. It's always fun to look at a user interface and finds what's wrong and what can be improved. In some way, it reminded me of some usability reviews that the usability team was doing for various applications a few years ago. That's an effort that we've been missing lately, and I'd love to see someone revive this!

The work on the new control center seems to be moving along nicely. We should see the results in the next development cycle; don't be afraid to help Thomas if you're interested in this!

While discussing preferences, and removing some of them that we think most (as in a huge percentage) people don't use, we mentioned the fact that when we remove some settings from the various configuration tools, a lot of people get unhappy, to say the least. This is understandable, but we also always pointed out that it should be easy to write a small tool to enable people to change those settings graphically again. That never happened, but we'd like to avoid further unhappiness. This is how the idea of GNOME Plumbing was born. And I foolishly proposed to implement this.

It was funny to see Garrett breaking his openSUSE installation. Except that it shouldn't break this way when using GNOME:Factory on 11.2. Oops.

I had a good chat with Jon about GNOME 3. There's so much we can deliver during the whole GNOME 3.x cycle... We're focusing on 3.0 right now, but we need to prepare the following releases too. It was motivating to get reminded of the various areas we should explore, and motivation is something that was most welcome :-)

It was good to catch up with Lucas, just a few days before Julia magically appeared :-) He's still one of my heroes.

On Friday morning, Bastien told to Mairin, Garrett, Jakub and Hylke: okay, you want tools for designers; we're a bunch of hackers here, but we need you to design the tools you need. This resulted in a good discussion. Except that now, we really need some people to sit down and implement this. I guess this could be an interesting Summer of Code project!

Seeing Willie get hopes for usable accessibility support in GNOME Shell was a real pleasure. It's been a hard topic for months, and knowing that there might be some light at the end of the tunnel is already good news.

Matthew invited me to a card sorting session about settings and how to group them. It was a new experience for me, and seeing someone struggle to organize settings was eye-opening: I got the feeling that even with just one person doing this seriously, we can improve the overall experience for many users. I'm intrigued how usability people deal with different people having conflicting behaviors, though.

Many thanks to Canonical and Google for sponsoring this hackfest, and also thanks to Klaas and Novell for letting me go on a short notice :-)

I have high hopes that putting all those designers and usability people in one room together during one week will also make the GNOME Usability team move forward again. Usability is an essential part of our DNA, but we've been slowing down our efforts there, instead of accelerating as we should have done. This hackfest should put us back on track!

Comments

1. HtH [08/03/2010@06:09]

Hi Vincent,

I made a comment on the following site (http://www.hadess.net/2010/02/were-...) with some suggestions for the preferences UI(Gnome Plumbing, I believe is the working name). I didn't get a response, and I wasn't able to find an appropriate place to solicit you. Luckily (for me!) your blog came up in Gnome Planet while this was still fresh in my mind. So, while I know it is very early for Gnome Plumbing, I'd like to make the following observations (this is an edited version of the aforementioned comment):

While I like GConf, I understand the need for something like this, but this design, while pretty, doesn't seem like it will scale well. In the mock-up there are only a few applications shown in the left frame. On a real system, you could have a very, very long scroll through all the apps that are installed. It makes more sense to use some sort of categorization similar to how the main menu is structured. While boring, it is very easy for the user to navigate and it would scale very well with a careful use of collapsing fields like an accordion. It would also be nice looking and be in keeping with the look of Gnome Shell (I'm assuming this is more of a 3+ thing anyway). Lastly, later in the week Calum Benson posted that "pattern" was probably going to be a part of Gnome 3. Following the link he gave I found this http://www.welie.com/patterns/showP... . Well, that seemed like a nice coincidence to me:)
If you are at all interested in something like I mentioned, I could send you a quick mock-up. As for actually implementing it, I'm really not sure it could be done using gtk widgets. However, since this is for Gnome 3, and hardware acceleration is presumed (thus far), that may not be a problem. Regardless, I am no expert on gtk, so maybe there are some hacks that could be done that wouldn't require Clutter?

Best/Liam

2. Vincent [08/03/2010@11:09]

@HtH: I'm not that worried about scaling right now, for various reasons. First, I'm unsure that that many applications will ship settings to be tweaked this way, but maybe I'm wrong :-) Second, we can definitely use a group system if needed. And third, I expect most people to use the search to find the settings they're interested in.

Of course, we'll see how it goes.

3. Captain Magnus [08/03/2010@11:09]

I think it makes a lot of sense to hide settings that the "common" user will most likely never need to alter. *But*, all the other settings that are "hidden" should be documented in one place so that it's easy for developers of any power settings tool can pick it up and incorporate in their utilities.
In fact, I think this should be mandatory!

Cheers,
Magnus

4. Vincent [08/03/2010@11:12]

@Captain Magnus: well, arguably, all settings are already documented in gconf, with the schemas. We probably don't want to offer all settings available in gconf in such an application (some of them might be needed in some really special use cases only), and it's also unclear if all power settings tools would want to list the same set of settings. Some might prefer to focus on a specific use case. For example, pessulus could be considered as such an application, but focused on lockdwon.

5. Stoffe [08/03/2010@14:18]

If the window manager could do tabs and split views in a reasonable manner, all of those problems would disappear for a lot of apps. I'm somewhat addicted to both, but don't think they necessarily belong inside each application - not to mention that tabs and split views could then be used for *different* apps.

6. HtH [08/03/2010@19:23]

@Vincent: I may be mistaken, but isn't this intended to be the true outward face of Gconf? in your follow-up to Captain Morgan you say that you don't want to necessarily offer all the gconf settings. Okay, but that still means that pretty much every app has an entry.
Your point about search is well taken. My only issue with that is that people new to Linux may not know the name of the application for which they are searching. Since this tool seems to be focused on that group, along with the non-geek crowd as well, I think, it makes sense (to me at least:) to give them a nice, familiar interface with the grouping of applications.
At any rate, I very much appreciate your response. I've been using Linux for awhile now, but only very recently have I felt compelled to take an interest in improving the experience. I would like to help, but thus far the reception has been somewhat cold, even hostile at times.

Thanks,
Liam

7. Vincent [08/03/2010@21:58]

@HtH: now, it's not intended to be the true outward face of gconf. It's intended to be used for settings that can be used by a large enough group of users.

And you don't have to know the name of an application to search. Searching for "browser" should give you the relevant settings for web browsers (say, firefox and epiphany) as well as settings from other applications that are related to a browser ("browser mode" in nautilus?), for example.

8. m [09/03/2010@18:18]

"removing some of them that we think most (as in a huge percentage) people don't use"

Why don't _measure_ it? Like Banshee guys do: http://gburt.blogspot.com/2010/03/b...

9. Vincent [09/03/2010@18:25]

@m: that'd be nice, we just need to find people who will have time to do it with our current infrastructure, though :-) That's something we could possibly integrate as part of Plumbing.

10. HtH [12/03/2010@04:31]

@Vincent: I understand your points, but is it a good idea to have multiple entry points to what is, essentially, the same functionality? My understanding of Gnome 3 is that it is supposed to present a more unified environment, with less entry points.
Gnome Plumbing, as it stands now, seems more like a Gnome 2 presentation, though nicer looking than most. A Gnome 3 version seems like it should be the one place that everyone would go to for advanced settings (or use the commandline gconftool). Assuming the gconf-editor gets updated with Gnome 3, then there would be three different ways of doing essentially the same thing. Excluding specific options/settings from GP and assuming that gconf-editor is not included, you're relying on the developer to make the decision for what options should be exposed to those who should already be advanced users. That leaves the user to discover gconftool and then wade through documentation to find the settings that aren't included in GP. IMHO, that seems far beyond what should be a relatively minor task.

Best/Liam

my blog

Main Menu

Calendar

Search

Last Comments

Links

Feeds

Monday 8 March 2010

Tidbits from the Usability Hackfest

Comments

Comments are temporarily disabled.

by Vincent