Re: [Usability] Mezzo interface [was Re: nautilus ... ]

[Diego Moya <turingt gmail com>]

On 14/07/05, Alan Horkan <horkana maths tcd ie> wrote:
> 
> On Thu, 14 Jul 2005, Diego Moya wrote:
> 
> > "In pursuit of desktop evolution: User problems and practices with
> > modern desktop systems." (Ravasio 2004)
> >
> > Link: (might require access from an academic institution to read it)
> > http://scholar.google.com/scholar?hl=en&lr=&q=+Ravasio%3A+In+pursuit+of+desktop+evolution&btnG=Search
> 
> /me files link away for later reading
> thanks.
> 

For the sake of convenience I will copy here some of the tastier bits
of that document. Note that some of the problems described are being
finally addressed in modern desktops with tools lice Apple's Spotlight
and infrastructure like Beagle (but this one still lacks a good user
interface).



-----
4.2 Problems Identified
Based on the interviewees' statements, the problems they encountered can
roughly be separated into two main areas: those associated with the
user interface and those associated with the underlying system.


-- Problems concerning the user interface:
   (1) The use of metaphors offers the possibility to facilitate
system accessibility to users with varying skill levels. However, with
respect to the desktop metaphor as found in current systems, this has
not proven to be the case.
       Low and even medium skilled users rarely knew that documents
can actually be placed on the desktop itself. Only with increasing
experience, by observation and copying others was this knowledge
gradually acquired.

   (2) The screen plane was regularly misused by the system. Newly
installed programs tended to automatically add short cuts to the
screen plane. Even medium skilled users did not dare to throw the
short cuts directly to the bin, as they thought the respective program
might be disposed as well. Consequently, users preferred to feel
confused because of the icons on the screen rather than to take the
risk of having a program disappear `magically' and entirely from their
computer.

   (3) Low and medium skilled users expressed irritation due to the
functional similarities between the file system's user interface (File
Explorer in Windows, Finder in MacOS) and the screen plane's user
interface (the actual desktop). One interviewee explicitly expressed
his irritation, saying that he perceived them as two totally different
entities and did not wish their functions to be almost entirely
identical.

   (4) The user interface still acts too intrusively: One user claimed
details such as "those flying papers", during the copying process to
be extremely irritating, since he would usually be busy working. For
this concrete case he thought it preferable to have a simple status
indication that would allow him to know that the system is active and
completing the task.
       
-- Problems concerned with the underlying system:
   (1) The systematic separation of files, emails and bookmarks was
determined by three users to be inconvenient for their work. From
their points of view, all their pieces of data formed one single body
of information and the existing separation only complicated procedures
like data backup, switching from old to new computers, and even
searching for a specific piece of information. They also noted that
this separation led to unwanted redundancy in the various storage
locations.

(2) Remarks and short-term notes: Sometimes files, emails and
bookmarks are simply not enough to deal with the requirements of daily
work. Users confronted the challenging task of adding and managing
remarks and short term notes: The interviewees felt the need for a
comfortable, flexible, easy-to-use method for adding notes and remarks
to documents, much akin to the physical world practise of using sticky
notes. None of the available electronic solutions (e.g. Adobe Acrobat
Annotations [Adobe 2001] or 3M Post-it Software notes [3M 2003]) were
rated either useful or practical.
Two users  even misused their mailboxes as containers for short-term
notes so as to store data electronically and safely in an easily
accessible place

(3) Link together resources that are the same or belong together: Most
interviewees expressed the need to have their information linked
together (e.g. article author and respective address book entry, or
citation and cited article, etc.) and in general, to have more
content-based and context-based access to their information. To
compensate for this system `defect', one of the highly skilled users
presented us with a surprising, though pragmatic solution: He decided
to interconnect the information he owned using a MS Word document and
hyperlinks. Based on the thematic structure inspired by collapsing
menus, he then linked internal references (chapters, summaries,
pictures, author names, citations etc.) with external, mainly locally
stored resources (e.g. address book, web pages, PDF or PS files).
However, this user was still not satisfied with his approach, as it
only allowed for one way linking from reference to resource, but not
the other way around. "It's still not what I am really looking for."

(4) Classification and gaining overview implied considerable effort:
Our subjects complained about the effort connected with structuring
and ordering information manually. They thought this was a task that
their computer could address, though naturally they also wanted to be
able to adjust the result if they did not agree with it.
    Users agreed that an ideal system would offer an enhanced overview
of their data--far better than anything they had experienced to date.
The directory tree was considered too complicated to navigate and not
really helpful in depicting a global overview.


4.3 Conclusions on the Classification of Documents
Our study has shown that there are indeed a number of concrete
functional gaps that will have to be filled in order to make personal
computer systems better suited to users' requirements. These
requirements can be organized into three groups:

-- Properly separate information belonging to users and to the system.
Generally, today's systems do not incorporate a proper separation of
system-owned data and user-owned data. For instance, the `My
Documents' folder in Windows is hidden somewhere in the system's
intestines (namely under: C:\documents_and_settings\user_name\)
without any easily understandable reason for this to an average user.
Similarly, the screen real estate is misused as it is decidedly a
user-owned area, yet is used to automatically store, for example,
program short cuts.

-- Provide small but potentially extremely helpful tools to manage
information. Our users only explicitly identified a potentional
annotation tool, but others might be thought of as well.

-- Integrate rather than separate information. The present
hierarchical file system does not allow for many of the `networked'
information access procedures that seem so natural to us. As an
example, consider the following task: "I need to write an email to all
people who formed part of the organizing committee for event X in
1996." How would one perform such a task using today's system and why?

   In our study we were able to verify the existence of the three
generic types of information (temporal, working, archived) [Barreau
and Nardi 1995]. Our evidence further suggests that all the efforts
invested in organizing, naming and maintaining the hierarchical file
system structure are aimed at (1) engraving the information's content
and context into the system, and (2) providing an overview in a single
glance, without having to access information until it is really needed
(see also Shneiderman's Visual Information Seeking Mantra [Shneiderman
1996]).