Advertisements

Tag Archives: Data

A Day With Edward Tufte

 

Graphic of Napolean's March

One of the more iconic images Professor Tufte uses in his presentations. I have a mounted, autographed poster of this one.

 If you create reports, presentations, info graphics, or are in any way involved with presenting data of any sort, I hope you’ve heard of Edward Tufte. Even better if you’ve heard of his work, especially what I believe is his seminal book, “The Visual Display of Quantitative Information.”

As part of my job at Rocketdyne, I was privileged to attend an all-day seminar of his in Los Angeles in the Spring of 2007 or ’08. Upon my return, I wrote down some notes and impressions for my colleague who paid for the day. There’s some really good stuff in here. As a knowledge management professional, I’m a bit chagrined it’s taken me this long to share it. I truly hope someone finds Tufte’s words useful. 

Bill:

Here’s a quick recap of Edward Tufte’s presentation last Thursday. What I did, for the most part, was enter points he made as numbered bullets. Therefore, I’ll do the same here with the addition of some extra comments if I feel they are necessary.

1. Professor Tufte refers to the nature of the work he does as “escaping flatland”. He believes dimensionality is extremely important when using visualization to represent quantitative data.

2. Another aspect of visually presenting data which he emphasizes is data density, i.e. resolution. He repeatedly stressed the need to drive for greater and greater resolution when presenting data.

3. With respect to items such as run charts, histograms, etc., he believes it is far better to label the data directly, avoiding the use of keys, which he feels are distracting.

4. He presented a copy of Euclid’s Elements, which included many “pop-up” graphics used to illustrate his points. The copy of the book he had an assistant bring around (wearing white gloves) for us to view is 432 years old. It was awesome just to see it. He refers to these pop-ups as the “brute force” method of escaping flatland.

5. A key point he stressed is to enforce visual comparisons. The terms he used (should sound familiar) were, “it depends” and “compared to what?”.

6. The visual representation of data should show mechanism, process, or dynamics, i.e. they should present causality as an aide to understanding and clarity.

7. He also stressed the importance of showing more than 1 or 2 variables when preparing a chart.

8. Presentations must be content driven, i.e. they must embody the three elements of quality, relevance, and integrity. Integrity was a big theme of his and one I don’t believe most of us would find fault with.

9. Design can’t rescue failed content, which he referred to as “chart junk”. This is another point which relates to integrity, and one which he continually stressed throughout his presentation.

10. Whether it’s drawing or words, it’s all information. Don’t be afraid to use words to make your point.

11. I’m not entirely certain of what he meant by this point, but what I wrote down was the following: “Better to show info adjacent in space as opposed to stacked in time.”

12. He stressed that you should use small multiples, i.e. strive for high resolution of the data.

13. Another point which he used to continue driving home the importance of integrity was to show the whole data set. At the same time he stressed that one need not show the zero point, i.e. context is what’s important in making a useful, accurate presentation.

14. Detail does not mean clutter. If you can’t present your data in sufficient enough detail without making it difficult to understand, rethink your design; it’s probably faulty.

15. When presenting data always normalize, adjust, and compensate to provide greater clarity and integrity. The example he gave for this involved a situation where it was impossible to know the real changes in costs of consumer items without taking into consideration the rate of inflation over a period of time. Absent this adjustment, the changes appeared to be far greater than they actually were.

16. Perhaps this next point was specific to financial charts, but it seems appropriate for many others. Don’t trust displays which have no explanatory footnotes. Generally speaking, Tufte believes one should annotate everything. His philosophy appears to be to always err on the side of accuracy and completeness (see integrity).

17. He made a point of explaining the human mind’s tendency to remember only the most recent (recency bias) data it perceives. I don’t remember the exact context in which this statement was made, but I think it is related to Ed Maher’s assertion that we tend to focus on the out-of-family (I can’t remember the exact phrase he used) experiences rather than the steady state.

18. He used a word I thought was interesting to describe people who create fancy charts which don’t actually say much – “chartoonist”.

After going into some detail regarding how the Challenger disaster occurred or, more accurately, how it was allowed to happen, he suggested there were three moral lessons to be learned from the experience. He posed these lessons in the form of three questions one must ask oneself when producing information of this nature.

1. Where is the causality?

2. Is all relevant data included?

3. What do I really need to see if I’m going to decide this?

He guaranteed if these three questions were adequately addressed, the chance of getting the decision right were greatly increased.

He then went on to lay out a list of rules for presentations, as follows:

1. Get their attention (he gave an example of what he called the “stumblebum” technique, where a presenter purposely made a mistake – which the audience was more than happy to point out – in order to insure everyone was paying attention (presumably to see if they could catch him again; which they never did.) He made a point of suggesting this probably wasn’t the best technique, unless you’re really good.

2. Never apologize – don’t tell the audience how you didn’t sleep well the night before, etc.

3. PGP – Start with the particular, move to the general, return to the particular.

4. Give everyone at least one piece of paper; something tangible they can leave the room with.

5. Respect your audience’s intelligence.

6. Don’t just read from your charts.

7. Forget K.I.S.S. – Be thorough and accurate, not simple and vague.

8. He stressed the importance of humor, something he was excellent at. He did caution appropriate use (duh?).

9. If you believe what you’re presenting, make sure the audience knows it.

10. Finish early

His final points to improving one’s presentations were directed to the presenter and the presentation, respectively. The first point was to practice or rehearse so the presentation goes smoothly and you are able to get through it without stumbling or going over your allotted time. The second was to have better, stronger content.

Professor Tufte’s presentation was extremely engaging, from my point of view. He knew his stuff and made it interesting, fun, and funny. I confirmed that most of what he discussed is contained in one or more of the three books I took from the seminar, and I’m looking forward to reading again what I think I learned from him. Much of what he had to say was common sense, which I have encountered previously from the years I’ve spent putting together presentations. Nevertheless, I believe he had a great deal to offer which will ultimately improve my ability to present information, whether in a briefing or on a web site. I really enjoyed seeing and listening to him. Thanks for the opportunity.

Rick

Advertisements

Chasing That Elusive Health & Fitness Goal

I have long wanted to get back to my original weight, 7 lbs. 9 oz. but I’m finding it difficult. Regardless, recently I purchased a Fitbit Charge HR digital tracker to monitor the exercise I get and, shortly afterward, a Fitbit Aria electronic scale. I won’t say I’m actually part of the #QuantifiedSelf movement, but I do like data and find they help me achieve goals by showing me how I’m doing and the consequences of not following the steps needed to accomplish them.

 

Fitbit Aria

My New BFF


Last Friday marked two very important milestones in my quest to get in better shape and, more importantly, to reach a point where I can either stop taking the two maintenance drugs I’ve been on for quite some time (for essential hypertension and type II diabetes). I test my blood sugar at least every morning and Friday, for the first time in my memory of the last over 15 years, it was under 90 upon arising. This is very unusual for me as I have always experienced an early morning spike in my readings.

I’m also weighing myself each morning as soon as I get up. The Aria scale measures weight and body fat percentage. After I stand on the scale and it settles, it shows me my weight, body fat %, and my initials (it will recognize up to eight people) twice, then syncs the data via our wifi to my Fitbit account. This morning I dipped below 180 lbs. for the first time in decades.

Over the weekend I indulged a wee bit and this morning my weight was just over 180, but my blood sugar was 89. My average before meal reading is now about 110, an amazing difference from what I’m used to. Last time I had an A1C test, I had dropped below the threshold of 7.0 and I’m quite certain it will be even lower this time. I suppose I could have achieved this a long time ago, but I didn’t. Better late than never, eh?


Serendipity Runs Circles Around Causality

Funny how some things seem — given enough time — to come full circle. Although I have always seen patterns and complexity, as well as the intricacies of their interplay, I wasn’t introduced to the concept of Systems Thinking until I worked on the Space Shuttle Main Engine program at (what was then) Rockwell International’s Rocketdyne division. That introduction included being exposed to the thinking of Dr. Russell Ackoff, a recognized authority in the field. I was fortunate enough to spend some time with him, twice in Philadelphia, prior to his death in October of 2009.

Shortly after retiring from Rocketdyne in 2010, I was introduced to Dr. Lorien Pratt of Quantellia, LLC, who showed me a tool her organization had developed called World Modeler. I was excited at what I saw and hopeful I could somehow become involved with Dr. Pratt and her team. However, that was not to be at the time. I was provided the opportunity to more thoroughly investigate the tool, but I had made a conscious choice to refamiliarize myself with Apple products (after over two decades of living in the PC, DOS, and Windows environment) and World Modeler was not written to be run on a Mac. Furthermore, the PC laptop I had wasn’t powerful enough to do the math and drive the graphics required for running models in real time. I was hosed.

Decision Intelligence Technologies

Decision Intelligence Technologies

Finally, about six months ago I was contacted by Dr. Pratt, who asked me if I wanted to assist in writing a paper that described an effort in which they were involved with the Carter Center. I enthusiastically said “Yes!” I’ve done a couple of other things with Quantellia since then but, beginning a few weeks ago, I took on an entirely new and (for me) exciting role as a referral partner.

Right now I’m spending a fair amount of time learning Decision Science in general, and the process and tools Quantellia uses to help organizations understand complex interrelationships and make better decisions based on that understanding. As I’m doing this I watch videos, read blogs and articles, look for original research, and work on presentations that will help me educate others in this important approach to business and organizational operations.

So . . . here’s the full circle part. As I’m looking for definitions, or explanations, of Decision Science and its origins, I Google the term. The first two hits I get are to The Decision Sciences Institute and to Carnegie Mellon University’s Department of Social and Decision Sciences. The third link is to a Wikipedia article on that same Department. In that article, there’s a link to Decision Science, specifically. However, it redirects to an article on Operations Research, which is where Systems Thinking originated. At the bottom of the page is a list of researchers under the heading “See Also”. One of the researchers, unsurprisingly, is Russell L. Ackoff. To me, that’s a combination of serendipity and years of working on better understanding how an understanding of systems can work to the benefit of any organization; actually, anyone.

I’ll be writing a lot more about Decision Science, including my understanding of some of its constituent parts, Decision Intelligence, Decision Engineering, Decision Modeling, and the power and value of our tools, World Modeler and DEEPM (Decision Engineering for Enterprise Project Management). I hope I will be able to clearly explain what it is we have to offer and, more importantly, what everyone has to gain by understanding it. The value exists independently of me or even Quantellia. We’ve just been at it for a while and can apply and employ the discipline both efficiently and effectively. Stay tuned.


Intertwingled in Plain Sight

Intertwingled

All Things Are Ultimately Intertwingled

I’m going to continue on a theme from my 4th of July entry, where I kind of resurrected an old post of mine from Content Management Connection. This time, however, it’s not a post of mine but that of a friend, Greg Lloyd – President and co-founder of Traction Software, Inc.

There are two terms I remember from when I first read Greg’s post – originally published on July 5, 2010 – which have helped me understand what I expect from the application of knowledge management and social business (formerly Enterprise 2.0 © ) design concepts and tools. These two terms also help me describe several of the most important attributes and indications of a well functioning, successful organization or group. They are “intertwingled” and “observable work”.

As a knowledge management professional (hemidemisemiretired) my long-standing and overarching goal has been to help people (and their organizations) improve on their ability to make sense of the huge amount of data that flows from their work. Doing so requires consideration of both macro-environmental factors and micro-environmental factors. For me, intertwingle describes the macro environment and observable work is what helps the micro environment to thrive. Let me very briefly explain why I believe this. Then I’ll send you off to Greg’s wonderful post where he explains it far better than I am capable of doing.

I frequently use the term “systems thinking” to describe what I see as an ongoing process of understanding that recognizes the interconnection, as well as the interdependency, of . . . well . . . everything. Useful systems thinking also requires the ability to see boundary conditions in pursuit of knowledge, but keeps the systemic nature of all things in mind when considering how they work. The word ‘intertwingle” seems to succinctly embody what I just spent a paragraph attempting to explain; probably not very well. 😦

“Observable work”, on the other hand, evokes a vision of people communicating with each other and the data and information essential to the smooth functioning of the work they do. It promises not necessarily the disappearance of silos, but does suggest making those silos – and the varying and very real relationships they have with each other – more transparent and discernible.

There’s much, much more that flows from these two concepts but, since I have no intention of rewriting that which has already been published, I urge you to read Greg’s post. If you have the time and the inclination, you may want to follow some of the numerous links he provides that serve to further define and illustrate these two concepts. Think of it as a quest to find the social business/knowledge management version of the Higgs Boson particle or, at least, the Gluon.  Here’s the link.


Knowledge Management Ain’t Actually Going Anywhere

I completely forgot I had posted this over a year and a half ago. I never actually posted it here, but did post about it and provided a link to it at Content Management Connection. Despite the passage of time since I did post it, I don’t really think much has changed, but I’ll let you be the judge of that.

PS – Click here to see an up-to-date graph and some regional data as well from Google Trends.


Google Trends Graph

Knowledge Management vs. Social Media Searches via Google

As a result of two tweets I just read; one from @SameerPatel and the other from @ralphmercer, I wanted to get a thought down before it recedes forever into the darkest corners of my brain, where I know I will feel the remnants of its presence, but will also never be able to fully recall it.

Based on something Sameer said I went to Google Trends and searched on the terms “Knowledge Management” and “Social Media”. In the past almost two years, with the exception of a large drop at the end of 2009, and a slight dip at what looks like the end of June in 2010, Social Media searches have been steadily increasing. During that same time period, searches for Knowledge Management – which are now less than a fifth of the searches for Social Media have remained arguably steady, with perhaps a bit of a continuous waning.

I suppose some would suggest this portends the eventual death of KM, but I really don’t think that true . . . or even possible. KM has always been based on the belief that we humans are unique in our ability to pass knowledge on to others, as well as to collectively create new knowledge and retain it for future use.

As I had suggested to Ralph, and what he was kind enough to point out in his tweet, is the reality that it’s “very expensive to reacquire knowledge”. This isn’t something anybody wants to do, anymore than they want to produce re-work or scrap. Yet people seem to be mulling over the viability of KM for the future.

I think the reality is two-fold. First, the need for sharing and re-using knowledge or information continues as strong as it’s ever been. What it’s called is of little consequence and, if KM has gotten a bad rep, then let’s move on and call it something else.

Second, I believe a lot of what we mean when we refer to social media is actually the next iteration of KM, insofar as it enhances collaboration, sharing, finding out what others are doing, etc., as well as captures and makes available collective knowledge and wisdom.

So, what do you think? Has KM run its course, or is it just taking on a new “identity” in the form of social media and (something I don’t think I mentioned above) Enterprise 2.0?


How Networked Science is Stretching Our Vision

Making Sense of it All

Making Sense of it All

My original intent for this blog was something far different than it’s become. I don’t think it’s a problem, as I seem to be morphing my approach into something that can easily accommodate that original intent. In case you aren’t aware of what I wanted to do when I began this little journey, I have described it somewhat here. I intend on updating that page periodically to keep up with the developments and changes as they occur (or, hopefully, shortly thereafter). This continues to be a work-in-process and I think that’s how I want it.

I have had a deep love and respect for the concept of Systems Thinking a good part of my adult life. As a young man I didn’t even know it was something people studied or wrote about; just that it seemed to be a useful way to look at the world and try to make sense of it. Recognizing the systemic nature of things and seeing the interrelationship (no matter how distant or tenuous) between them can, in my opinion, make them far more intelligible while increasing the odds of understanding consequences and why certain things happen.

Today I came across a wonderful article on Facebook from The Atlantic, through a post by John Hagel of the Deloitte Center for the Edge. I’m not “friends” with John, though I have sent him a request. In the meantime (and I assume he will likely ignore me) I do “subscribe” to his public updates. He shares some truly fascinating and interesting information.

The article is entitled “To Know, but Not Understand: David Weinberger on Science and Big Data” and he discusses and explains how the prolific growth of data, information, storage capabilities, and computing power is facilitating the understanding of large-scale or highly complex systems, despite their being beyond our ken as mere human beings. He points out that, despite our limitations as individuals to understand why some things work as they do, the growth of networked science is providing us with a capacity for making use of this data, information, and knowledge. I found it truly fascinating and want to share it here. If you have 10 – 15 minutes, I highly recommend you take the time to read it. Here’s the link.


%d bloggers like this: