Re: GNOME 2.23 Schedule

[Olav Vitters <olav bkor dhs org>]

On Tue, Mar 18, 2008 at 07:58:59PM +0200, Felipe Contreras wrote:
> >  To summarize this, GNOME ain't Windows and we ain't Microsoft. We
> >  don't and probably will never have budget to have huge marketing
> >  studies about what users want. Microsoft thosed tons of money to
> >  investigate new interface for Office - and it still have made lot of
> >  users mad. So I suggest not to try to waste our time to this.
> 
> There are many ways to improve usability[1], not all of them are expensive:

Expensive as in money or time? If you want to start such a investigation
 -- go ahead. E.g. Novell has a lot of usability studies. There is a
website where you can see how to setup a usability testing lab, etc.

My response to 'GNOME/someone should do this': go ahead and do it
(completely honest answer -- I value action).

> Focus Group:
> A monthly online focus groups doesn't cost a thing. Some mortal "user"
> can be in charge of gathering conclusions so no developer is
> disturbed.

I think you're thinking in monetary cost.. because you have to make sure
that the group is a good representation of your target group. That takes
time and effort (both in selecting the group plus convincing GNOME
contributors of that fact).

> Usability Testing:
> Put your GNOME desktop in front on some user, watch him and write down
> his problems. Again, no developer needs to do that.

Already being done. If you want to do more of that, again: go right
ahead. I'd welcome that (but make sure of the target group).

> Surveys:
> Just ask your users. Some web developer(s) would need to work on this.

Online surveys are really not what you want.. you don't know who
responded (online answers = questionable quality). Plus you might get
very strange results because only a small percentage of your users
responded, while those are e.g. out of your target group.

> </snip>
> 
> 
> The problem is that developers don't like users; who would scratch

This is a very strong statement. I don't agree, and I am a developer. I
just ignore certain people, but that doesn't mean I ignore users.

> somebody else's itch? That's not a big deal in a lot of open source
> projects, because developer equals to user, but in a Desktop
> Environment the diversity of users is overwhelming.
> 
> So not all the issues get visibility, only the ones of patient users,
> at least enough to find out the proper ways to report them. There's a
> lot of users without that amount of patience.

What is bad about that? GNOME couldn't handle if every user provided
feedback -- we would be totally overwhelmed (and ehr, Bug-Buddy is
curently having that effect). Look in Bugzilla, stuff is often reported
multiple times (bugs, feature requests, usability problems,
documentation issues, etc). Eventually it'll get noticed .

> What I'm talking about here is the Product P in marketing; if you want
> your product to sell, you need to make it appealing to your market.

GNOME is not like fast moving consumer goods. Even then -- a good survey
is not easy to do. You need to think about the questions (lots of
surveys have bad questions / terrible multiple choice answers). But
still, go ahead -- prepare the survey. No problem if you e.g. ask here
or on the usability list for feedback.

Don't forget the analysis afterwards.. check for possible conflicting
outcomes, etc (perhaps unexpected answers due to the wording of a
question.. or some implied assumption).

> If the target market is devoted users who would fill bug reports and
> possibly attach patches, then GNOME is on the right track. If the

You assume those users dictate what GNOME does.. that is currently not
the case, so you can't assume if GNOME is on the right track or not.

> target market is 10% of the global desktop market then you need luck
> to achieve that goal without analyzing your true target market.

You started this offtopic (for the thread) with a suggestion 'a simple
poll', but I hope you see it isn't as simple as that. However, if you
want to do it, go ahead.

> [1] http://www.usability.gov/methods/

I'll read it. From a quick look I suggest to read the d-d-l and
usability archives. A lot of things have been discussed (e.g. usage of
personas). It also seems to prefer online stuff and seems made to
improve the usability of websites (IMO GNOME would need a different
approach).

But ehr, go ahead and start something.

-- 
Regards,
Olav