A few random thoughts...

Some issues come down to configuration. It can be hard for a single user to 
isolate such problems when they only have one system.  Also, where users and 
QA/dev have different system states, step-by-step procedures don't neccesarily 
work.  Many low-tech users may also be hesitant to "play" with their system to 
try to isolate the problem.

Virtual machine images have long been used by software developmers to run test 
cases against a known baseline.  It would be useful to have a standard 
downloadable image to run with VMWare Player, so that users can have a second 
configuration to compare with their own.  If they observe it working in the 
image and that it is not "just OpenOpenoffice", they can more confidently work 
to isolate the configuration causing the issue, which they can then then report 
against the baseline. 

Further, part of effectively isolating an issue is being able to reliably break 
a working system.  Being able to play without fear of breaking an existing 
setup would be useful.  

The way low-tech users explain problems is sometimes misleading.  They often 
summarise error messages, and remote controlling their PC to "see" the problem 
is often required.  The virtual matching image might include a standard utility 
to do annotated video recording.  This might first be processed by a community 
help forum.  

Further, (not that I've experienced this, but I was taught that...) usability 
labs record a user's keyboard-mouse-menu choices to perform some standard 
tasks. They run statistics on these to identify the dead-ends users end up in 
as they try to discover how to do a set task.  The image could be preconfigured 
with a standardised tool to generate such reports that a user can send in.  
This data could be made available to student research projects or for the 
community to organise.

Cheers, Ben


-----Original Message-----
From: Mike Hall [mailto:[EMAIL PROTECTED] 
Sent: Sunday, 9 April 2006 1:38 AM
To: [email protected]
Subject: RE: [qa-dev] A proposal to improve the bug entry page for "simple 
users"[Sophos Scanned]

-----Original Message-----
From: Andre Schnabel [mailto:[EMAIL PROTECTED]
Sent: 08 April 2006 11:58
To: [email protected]
Subject: Re: [qa-dev] A proposal to improve the bug entry page for "simple 
users"

 

Hi Mike,

 

 

> I think this is a very important issue. Coming from many years
professional

> involvement with PC application software, it often seems to me that

> OpenOffice has a too 'techie' outlook and, sometimes, priorities are 
> set
by

> what is interesting and sexy for all us techies rather than listening 
> to
the

> pleas of 'simple users' for some of the niggling but for them 
> important

> defects to be fixed. Anything that can be done to encourage these 
> users to

> make their views heard would be beneficial. After all, they are the 
> vast

> majority of users and, if OpenOffice is to make a real impact, 
> especially
in

> the developed world where cost is not such an issue, we need to make 
> these

> users comfortable. I'm not against increased functionality, I need 
> some

> myself, especially in Base, but from an OpenOffice 'business' 
> perspective

> perhaps the focus should more often be on useability rather than on

> functionality.

>   

 

I'd definately vote +1 if you would go on and establish a community 

driven usability project :-)

(Because I totally agree on your point, but it is a mess of work to get 

this coordinated)

 

André

 

Hi André and thanks for your positive response.

The principle of what you suggest is interesting, but I've not convinced of the 
need for a 'community project' - well I guess it depends what exactly you mean 
by a project.

Much of the information needed to improve usability is accessible, mainly in 
the issues database, though I agree it's a mess to get at it. The challenge 
though seems to be as much to influence the hearts and minds of those setting 
priorities so that more weight is placed on usability. 

As one possible practical way of going forward, perhaps the community could 
declare a forthcoming release (2.0.4?) as a 'usability release' and include in 
it a range of relevant issue resolutions. I'd certainly be willing to do some 
trawling to identify possible candidate issues. I guess that adding a comment 
something like 'please consider raising the priority of this issue to 2.0.x for 
usability reasons' would be a sufficient prompt to the owners, leaving them in 
full control. I'd be tempted to focus primarily on Writer (because it is the 
most heavily used) though not to the complete exclusion of other components. 
Clearly some issues would be generic anyway.

Would it make sense to invent some kind of marker for usability, whether a flag 
or a key word so that over time these issues were easier to identify?

How do I judge whether or not there is sufficient support to make it worth 
putting in some effort? Is there a better way to approach usability? Is this 
the right place to canvas for collaborators?

Mike      

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to