On Feb 15, 2008, at 2:53 AM, grass-dev-request@lists.osgeo.org wrote:
Date: Fri, 15 Feb 2008 11:53:17 +0200
From: Wolf Bergenheim <wolf+grass@bergenheim.net>
Subject: Re: [GRASS-dev] GRASS in a usability review
To: Tim Michelsen <timmichelsen@gmx-topmail.de>
Cc: GRASS developers list <grass-dev@lists.osgeo.org>
Message-ID: <47B5610D.3060909@bergenheim.net>
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
On 13.02.2008 23:57, Tim Michelsen wrote:
Hello GRASS developers,
OpenUsabilty has opened a call for projects to participate the SoC like
usability review. The deadline is the 20.02.2008.
Yes! That would be a great opportunity, not to be missed. I think we
should definitely participate. What does that require from us? (In SoC
we needed to assign a Mentor) The user interfaces could use guidelines
from usability experts, to point out things which we might take for
granted, streamlining the UI and maybe something else.
I hope that my thoughts and idea is welcomed and look forward to discuss
this issue with you.
As I said I think we should use this opportunity as long as we have the
means. I'm always interested in talking about usability for GRASS. Hope
we get some more interest then in the past...
In theory, a usability study would be very helpful. But (big caveat), I worry about results from someone who doesn't understand GIS. This concern comes from several years of fielding well-meaning suggestions (often the same ones repeated) from newbys to GRASS about why don't we have x--when a GIS simply doesn't do x?
GIS doesn't operate like a word processor or spreadsheet. So it's difficult to evaluate how useable a GIS is unless you know what a GIS is supposed to do.
Within those parameters, I think a usability study is a good idea. I remember a year or so ago, a users made a very good suggestion that we not have menus more than 2 levels deep and that menu item names should be shorter. This was difficult to do, given GRASS's complexity. But having done it, I think it helped usability a lot.
In theory, a usability study would be very helpful. But (big caveat), I worry about results from someone who doesn't understand GIS. This concern comes from several years of fielding well-meaning suggestions (often the same ones repeated) from newbys to GRASS about why don't we have x--when a GIS simply doesn't do x?
I believe we can maybe educate the would be usability student, just like in SoC. The student's in the SoC last year didn't have any background in GIS, and I think the results were very good despite that. I don't think that we should be so scared of ignorance that we would pass this opportunity, it is simply a matter of education.
GIS doesn't operate like a word processor or spreadsheet. So it's difficult to evaluate how useable a GIS is unless you know what a GIS is supposed to do.
Of course, but some things are universal, like when you do X you can expect Y. And re-use of UI elements like color selection. Also things like button placement and functionality can be maybe considered. I'm in no way a usability expert, not even close, but I've heard many comments about how hard GRASS is to use, and I think that with a good usability design we can maybe lower the learning curve a bit.
Within those parameters, I think a usability study is a good idea. I remember a year or so ago, a users made a very good suggestion that we not have menus more than 2 levels deep and that menu item names should be shorter. This was difficult to do, given GRASS's complexity. But having done it, I think it helped usability a lot.
Yes, I agree. It is a lot better, but I believe it could be made even better
I remember a year or so ago, a users made a very good suggestion that
we not have menus more than 2 levels deep and that menu item names
should be shorter. This was difficult to do, given GRASS's
complexity. But having done it, I think it helped usability a lot.
from the "hey, life's a compromise, do the best you can" department:
but try and balance that against the "platoon rule"; each sublayer in
the hierarchy should have no more than 12 subordinates. Beyond that it
becomes overwhelming to keep track of and quickly navigate through.
I guess groups of modules separated by horizontal line breaks within a
single menu could help with the compromise.
I believe we can maybe educate the would be usability student, just like in SoC. The student's in the SoC last year didn't have any background in GIS, and I think the results were very good despite that. I don't think that we should be so scared of ignorance that we would pass this opportunity, it is simply a matter of education.
I may remind that the GIMP (one of the first FOSS programs with a GUI) has hugely benefited from the usability improvements. GIMP is a (semi-)professional that is also not as simple as spreadsheets...
Since all started in Berlin a number of German podcasts have already covered this initiative.
They are not graphic artists who are to beautify the icons and windows. They look at the program and check:
* did the community define proper goals where the GUI should develop to?
* does it meet the users needs?
* what are typical workflows?
* what needs to be improved to enable users to accomplish their work efficiently.
I know people who wrote their thesis using GRASS and later shifted "because of the GUI".
Of course, but some things are universal, like when you do X you can expect Y. And re-use of UI elements like color selection. Also things like button placement and functionality can be maybe considered. I'm in no way a usability expert, not even close, but I've heard many comments about how hard GRASS is to use, and I think that with a good usability design we can maybe lower the learning curve a bit.
Yes, this is what could need a improvement.
Yes, I agree. It is a lot better, but I believe it could be made even better
Since I could read at least 2 "Yes, let's try!" I would like to ask for another developer to give me a strong confirmation before I fill the application form.
I am not a strong developer, but i do give GIS trainings (2 days courses intro to GIS) with GRASS GIS a lot this year (3 done, about 6-7 to go until May, maybe more later this year, more advanced level on hydrological stuff maybe).
Since it is on Windows, and most students are more standard Windows users, they have difficulties with learning curve from 0 level to something.
Any improvement in the GUI and workflow will benefit adoption rates, I am sure about it.
Since I could read at least 2 "Yes, let's try!" I would like to ask for another developer to give me a strong confirmation before I fill the application form.
This needs to be done TODAY (19.02.2008).
Timmie,
I'd say go for it. If you are willing and able to serve as a technical contact even better. Everybody else seems to be too busy I might also be able to help out.
--Wolf
PS. You might want to update your address book for the mailinglist address.
On Feb 19, 2008 1:51 PM, Wolf Bergenheim <wolf+grass@bergenheim.net> wrote:
On 19.02.2008 02:06, Tim Michelsen wrote:
> Since I could read at least 2 "Yes, let's try!" I would like to ask for
> another developer to give me a strong confirmation before I fill the
> application form.
>
> This needs to be done TODAY (19.02.2008).
Timmie,
I'd say go for it. If you are willing and able to serve as a technical
contact even better. Everybody else seems to be too busy I might also
be able to help out.
I also think that this could become useful and interesting. "Know your
community..."
I might also be able to help out.