This website covers topics on knowledge management, personal effectiveness, theory of constraints, amongst other topics. Opinions expressed here are strictly those of the owner, Jack Vinson, and those of the commenters.

PIM articles from the ACM

The latest Communications of the ACM (vol. 49, no. 1, Jan 2006) has a great set of articles on Personal Information Management.  The topics range from the broad discussion of what PIM is and how supporting technologies might work, to specific issues associated with PIM in our world today (security, scalability, data design), to behavioral research of people interacting with PIM tools. 

What is personal information management?  In the classic sense, it is a means by which I maintain information that is important to me.  At minimum, this is a calendar and a contact list, possibly enhanced with a note pad and a task list: my (paper) day planner, for example.  With a little technology thrown at it, this becomes a wide variety of software applications.  And it begins to beg the question of how much personal information needs to be kept, since the digital life is almost infinitely recordable. 

Many of the articles use stories to describe how the tools will be used in capturing and reusing "stuff" in digital form.  Future PIM tools will, for example, cross link a photo I took with my GPS data at the time of the of the photo and the person in the photo, possibly even superimposing a collection of photos on a map of where I've taken them.  Or, in a more business-like setting, the tool might easily connect notes I've taken with websites viewed while writing the notes and phone calls I've made in the same time frame. 

Much of the inspiration for the articles on personal information management tools comes from Vannevar Bush and his "memex" as described in As We May Think (Atlantic Monthly, July 1945).  The Memex has been referenced many, many times over the years of reading about PIM and knowledge management and PKM.  Have a glance, if you aren't familiar.  The abstract gives you a few hints:

As Director of the Office of Scientific Research and Development, Dr. Vannevar Bush has coordinated the activities of some six thousand leading American scientists in the application of science to warfare. In this significant article he holds up an incentive for scientists when the fighting has ceased. He urges that men of science should then turn to the massive task of making more accessible our bewildering store of knowledge. For years inventions have extended man's physical powers rather than the powers of his mind. Trip hammers that multiply the fists, microscopes that sharpen the eye, and engines of destruction and detection are new results, but not the end results, of modern science. Now, says Dr. Bush, instruments are at hand which, if properly developed, will give man access to and command over the inherited knowledge of the ages. The perfection of these pacific instruments should be the first objective of our scientists as they emerge from their war work. Like Emerson's famous address of 1837 on "The American Scholar," this paper by Dr. Bush calls for a new relationship between thinking man and the sum of our knowledge.

Projects referenced: Email as Habitat, Keeping Found Things Found, MyLifeBits, Stuff I've Seen, Visualization and Interaction for Business and Entertainment (VIBE).

Below are some more detailed comments on each of the articles, plus pointers to where you might find more on the web.  Note: Articles from CACM are available only for members or from the print journal in your favorite (academic) library

]]><![CDATA[

In Digital memories in an era of ubiquitous computing and abundant storage (Mary Czerwinski, Douglas W. Gage, Jim Gemmell, Catherine C. Marshall, Manuel A. Pérez-Quiñonesis, Meredith M. Skeels, Tiziana Catarci) the authors discuss a number of aspects of dealing with the digital memories that come in from ubiquitous computing devices, such as body-mounted sensors and cameras. 

They present a useful 2x2 matrix that helps show how and where digital memories are used. One axis describes who records the information (me vs. others), while the second axis describes who uses the results (me vs. others).  Their depiction helps realize the massive variety of information that we collect and that is collected about us, from the baby books parents create for their kids to medical records to journals to photographs and beyond.

This paper and several others describe a range of needs that these tools support.  It might be a simple need to enhance our memory for future reference, or it might be that we need the digital memories for later reflection and analysis of what we've done.  Time management is a classic reason for PIM tools, of course, and enhancing it with digital memories will help guide future assessments of what is needed when (tasks, appointments, etc).  We might want to share what we've collected with other people.  And having a digital record can also help us deal with legal problems (proof of purchase) and security issues (video surveillance). 

They also talk about the research challenges that arise with the collection of information that comes from such a wide variety of sources: what is important (now and in the future), integrating multiple sources, interacting with the accumulated information, privacy and access, and sharing it easily.

Personal health information management (Wanda Pratt, Kenton Unruh, Andrea Civan, Meredith Skeels) covers the question around health information and unifying technologies that enable the patient to collect their medical information from (and share it with) medical providers and family members in appropriate ways.

Usable privacy and security for personal information management (Clare-Marie Karat, Carolyn Brodie, and John Karat) discusses the authors' efforts in considering how to develop a "policy workbench" to set privacy and security controls over personal information.  Personal information makes this problem especially difficult, given the vast quantities of stuff that we generate and save and the need for individuals to have control over content disposition.

In Searching to eliminate personal information management (Edward Cutrell, Susan T. Dumais, Jaime Teevan) the authors talk about the Microsoft Stuff I've Seen project in relation to the vision Vannevar Bush had for fast search capabilities within Memex.  In their experience with Stuff I've Seen, they saw that people made heavy use of the search feature and the metadata filters to focus on what they were trying to find.  They particularly highlighted the importance of people and time metadata.  The people metadata is important because much of the stuff I've seen is associated with people (they sent it to me, they wrote it, I sent it to them, they are in the picture, etc.), and it is frequently easy to find stuff that is linked to people.  Similarly with time, it is common to look for stuff based on when I remember seeing it.  The authors also acknowledge the importance of tools that are smart enough to bring information to me without my having to explicitly search for it.  Stuff I've Seen has become Microsoft's desktop search tool.

Keeping encountered information (Catherine C. Marshall, William Jones) gives an overview of issues the authors have encountered in the Keeping Found Things Found project at the U of Washington.  They touch on an important aspect of the "keep everything" philosophy: if the user has to make any effort to keep something, the interruption may create more confusion (from dropped tasks) than the future value provided by saving the encountered information.  That said, it's quite difficult to know what is going to be useful in the future.

The article Email in personal information management (Steve Whittaker, Victoria Bellotti, Jacek Gwizdka) argues that email platforms are the most sensible basis for personal information management tools because they are a familiar environment for most people.  The central model for their research is that the stream of email relates to task management activities for the user.  They highlight importance of the content of email and its metadata (who, when, context) as critical keys to linking information to ongoing tasks in work and personal life.  Victoria Bellotti is associated with the Email as Habitat project at PARC, which has coined the term "thrask" to connote a user-designated collection (thread) of email associated with a task.  This is separate from traditional email threads because the user's task does not always correlate with the complete thread for the group.

The authors of this paper also suggest a number of changes that will come about as email tools advance and grow their PIM capabilities.  They are particularly interested in the change from simply reading email to using the stream of email to inform the user's task management activities.  They also recognize the growing trends of instant messaging and other non-email collaboration technologies as somewhat supplanting the role of email, but they also highlight the unique nature of email as it relates to task management.

The article From PIM to GIM: personal information management in group contexts by Thomas Erickson is a quick introduction to the importance of looking at what happens when you shift away from the focus on the individual to look at the group(s) in which the individual participates.  The idea also links back to the first three papers in that personal information is not always created by the subject of the information, and in some cases they don't even own the information directly (credit history).

Addressing the basic problem of how to integrate data from many different sources of personal information, Data unification in personal information management (David R. Karger, William Jones), shows where the field is going.  Karger has worked on the Haystack Project, which provides a unified data environment for PIM.  Jones has worked on Universal Labeler (part of Keeping Found Things Found project), which seeks to extend existing tools with data to enable users to integrate data across the systems. 

No matter how the data and system integrate, this all needs to happen in the background, so that users can get on with the task at hand, whether that is working on their latest project or putting together a slide how of their latest vacation.

Reminding us just how difficult it is to create a "perfect" PIM tool, Diane Kelly writes about Evaluating personal information management behaviors and tools.  Not surprisingly, different people have different needs from PIM tools.  Even individuals have different needs at different times.  So coming up with consistent methods to study human behavior around the tools is a challenge.

Finally, an article that isn't formally part of the PIM collection of article, but is an excellent summary of what is happening in the space, MyLifeBits: a personal database for everything from Microsoft's Jim Gemmell, Gordon Bell, and Roger Lueder.  The authors do a great job of providing perspective on the MyLifeBits project, which has been running since at least 2000.  I particularly liked how the authors described the history of the project as they moved from thinking about document capture to realizing that ubiquitous computing means the capture of images, video, GPS data and much more.  Given the extent of the project it is also interesting that they are essentially studying and discussing their own interaction with the system.

One "aha" that came out of this piece was the comment that "more is better" in terms of deciding what to store.  "We never regret capturing; but we often regret not capturing more."  Why?  They refer to basic understanding of how humans work, and suggest that the more the system logs, the more chances that users will run across the right "memory hook" that will help them find what they need.  Of course, this requires a sophisticated tool that doesn't overwhelm users while given them the right level of control.

A question that comes up as I read through the descriptions of various approaches to PIM and user interface is where are things going for the next step up?  It is fascinating to get a peak into what will be happening soon in projects like MyLifeBits, Haystack, Keeping Found Things Found and others.  In a few years, the ideas and technology discussed here should be part of standard desktop toolkits.  What then?

Knowledge technology trap

Backlogs are a symptom