In a recent article called ”We Have Met the Enemy and He Is PowerPoint” by Elisabeth Bumiller there is a big outcry to stop using Powerpoint because it supposed to make us more stupid in decision-making. I agree and can just reiterate a quote from the top US Intelligence Official in Afghanistan, Maj Gen Michael Flynn in the report “FIXING INTEL: A BLUEPRINT FOR MAKING INTELLIGENCE RELEVANT IN AFGHANISTAN”:
“The format of intelligence products matters. Commanders who think PowerPoint storyboards and color-coded spreadsheets are adequate for describing the Afghan conflict and its complexities have some soul searching to do.”
These are quite hard words directed towards his commanders in ISAF and the US Component in Afghanistan but I think he is right. However, the underlying issue is a desire to simplify things which should not be simplified. Combine that with a lack of vision when it comes to tools support for higher level of military command. Basically the tools supposed to support that kind of planning are either general purpose tools like Microsoft Office or highly specialised military application which exists in their own stove-pipe.
Oversimplifications
With Powerpoint comes a method and that method mainly consists of boiling information down to single bullets. Perfect for fine tuned marketing messages that want to leave just a few critical words or terms in the heads of the recipient. Not that good for complex reasoning around complex issues like modern conflicts. Powerpoint sets out to convey a message when we instead should focus on creating situation focused on improving our understanding.
Static representations
Most Powerpoint presentations are very static in nature. They usually represent a manually crafted snapshot of a given situation which means that it can become outdated very quickly. As time goes on there are more and more static presentations that should be regularly updated but usually never are. Either they disappear in the file sharing if the organisation lacks an Enterprise Content Management system or there is no process on monitoring which presentations that need to be updated. Usually because all the traceability is lost from when they were being created. Some companies have implemented some dynamic areas in their presentations were for instance weekly sales figures are updated when the presentation opens but that is far from keeping track of where the orgins for each bullet, diagram and images are.
Laborintensive work
As described in the article there are quite a few junior officers that spend time collating information and transforming into presentations. To start with there is much to be done to support this kind of ”research work” where users are navigation and searching for relevant pieces of information. However, after the information has been collated the next part of the work starts which is to transform that presentations using a template of some kind. Decision-makers usually have an opinion of how they want their presentations set up so they recognize the structure of the information from time to time. Add to that the fact that most organisations have a graphical profile to adhere which suggests a common styling and formatting of the content. To me all this really calls for a more semi-automated way of compiling this information. I am not saying that all content can be templated, far from it, but where it is possible it would save lots of time. Hopefully time that could be spent thinking instead of searching and formatting in Powerpoint.
Lack of interactivity
Another problem of these static representations are that since they usually take hours to compile and the flexilbilty in the actual briefing situations is usually low. If the decision-maker suddenly asks to filter the information from another perspective in say a graph the unfortunate answer will be: ”We will get back to you in half-an-hour or so”. Not exactly the best conditions to inspire reflections that puts complex problems in a new light. Spotfire has even written a paper around that which is called ”Minority Reports – How a new form of data visualization promises to do away with the meetings we all know and loathe”. The ability to introduce dynamic data which is interactive can bring us a new enviroment for meetings, especially if we also have access to large multi-touch walls that invite more than one person to easily manipulate and interact with the data.
Format matters
The General is right, format matters. There is a need for several different formats of the same information. Maj Gen Flynn calls out for more products based on writing which allows people to follow a more complex reasoning. That tackles the simplification aspect of the problem. However, there is still a need to do things together in a room and handing out written reports in Times New Roman 12 points is not the answer. In fact we really need a revolution in terms of visualisation of all that information we have decided to store digitally. Especially since we are increasingly able to provide structure to unstructured information with metadata but also able to collect data with XML-based data structures. We really need more presentation and visualisation support to be able to work productively with our information. However, we need less Powerpoint because it is a very time-consuming way to do stuff which can be done much better with another set of tools. Multi-channel publishing is an establish concept in marketing areas which means that the same content can be repurposed for print, web, mobile phones and large digital signage screens. We need to think in a similar way when it comes to what we use Powerpoint for today. There is even a complete toolsets such as EMC Document Sciences which, surprise, is based on templates in order to do customized market communications where static content meets dynamic content from databases. In this case based around common design tools such as Adobe InDesign.
The Space Shuttle Columbia experience
One tragic example of when the use of Powerpoint was a contributing factor was the tragic loss of Space Shuttle Columbia. The Columbia Accident Investigation Board (CAIB) took the help of Professor Edward Tufte from Yale University to analyse the communication failure that in the end made NASA to not be aware of the seriousness of the foam strike. The board makes the following finding which is all in line with General Flynn’s observations:
At many points during its investigation, the Board was surprised to receive similar presentation slides from NASA officials in place of technical reports. The Board views the endemic use of PowerPoint briefing slides instead of technical papers as an illustration of the problematic methods of technical communication at NASA.
Tufte continues to make the argument that the low resolution of the Powerpoint slides used forces technical terms to be abbreviated and thus adding ambiguity and that the usual large font size in headline also forces shortening. He also notes that the typography and hiearchies provided by the bullet organisation also added confusion and that in the case of NASA some more advanced typographic feature to handle math and other technical formatting is needed.
During the Return to Flight work later on this was further emphasized with the following statement:
“Several members of the Task Group noted, as had CAIB before them, that many of the engineering packages brought before formal control boards were documented only in PowerPoint presentations,”
Unfortunately, this is something I can relate to in my line of business. The main form of documentation are a slide shows being emailed around. Since you know that they will be emailed around without You being there to talk to them I believe many add a lot more extra text in them which makes them into some kind of in-between creatures. Neither slides shows nor reports. At least these added words hopefully reduced ambiguity some degree. I have now started to record my presentations with my own voice to help mitigate that to some degree.
The Physical Resolution is usually to low
To further add to the Columbia findings I have serious issue with how briefing rooms are usually setup to today. They usually have only one projector with resolution between 1024×728 or 1280×1024. Many laptops today have widescreen formats on the screen which when used on “clone mode” makes the image of a 4:3 format projector looked really skewed. When projector handles widescreen formats especially with a higher resolution they are never used because:
- Users are given computers with sub-performance graphic cards that really don’t handle full HD (1920×1080) resolution.
- Users don’t know anything else but to “clone” their screen. What you see on the laptop is what you see on the projector. Thus in essence limiting the resolution on the projector to what ever the laptop handles. Again because users have been given cheap computers.
- The resolution has to be turned down from the highest one “because everything became too small to see”. The reason for this is that the physical screen size is too small which makes the projector sit too close and the actual pixels too small to see from most of the room.
Combine that with Powerpoint templates with big font sizes we have a situation which means that not a lot information can be displayed for us which I also think adds to the oversimplification issue of this issue.
Why the Afghan “Spaghetti image” is actually rather good
The NYT article contains an images from the Afghanistan conflict with hundreds of nodes being connected by arrows in different colors and this is given as an example of the problems of using Powerpoint. To start with I am not even sure that the image is made in Powerpoint, at least not from the beginning. I think a likely candidate instead is Consideo which is a MODELING tool not a PRESENTATION tool. The problems with that image is that when it enters the Powerpoint world it is static with no connections to underlying data. Imagine instead that that images was a dynamic and interactive visualizations of objects with relationships objects powering the lines. Metadata allows for filtering based on object and relationship attributes. Suddenly that images is just one of almost endless perspectives of the conflict. Imaging if all these nodes also are connected to underlying data such as reports and written analysis. Then it becomes easier even for an outsider to start understanding the image. We also need to understand that some visualizations are not intended for the decision-maker. Sometimes in order to understand them you need to have been there in the room most of the time so you understand how the discussions were. So this images is potentially rather good because it does not contain oversimplified bullets but instead is something you probably could stare at for hours while reflecting. However, it MUST NOT be an image that is manually updated in Powerpoint – it has to be a generated visualisation on top of databases.
Still valid for marketing
The almighty Master of presentations, Steve Jobs, who actually is using Apple Keynote instead of Powerpoint will most likely continue using that format. He delivers a very precise markeing message with slides that does not contain very much text at all. The rest of us who are not selling iPads need to start figuring out a smarter way to do business. Newer versions of ever more complex MS Powerpoint applications are simply not the answer. It is so general purpose that it doesn’t fit anyone any longer. At least if you care about your own time and data quality. It helps to some degree that both Keynote and Powerpoint use XML today – that means that the technical ability to use them as just a front-end is possible. The real issue has to do with information architecture and usage.
Conclusion
Oh, so how to do this, then? Use Enterprise Content Management systems to manage your content and move to an concept where content is handled in XML so it can be reused and repurposed while preserving tracability. Have a look at my other blog post around “The Information Continuum” to get an idea of how. Since we do store all of our information digitally there is a need for much more in terms of visualisation and presentation support tools – not less. However, we need to find a way to be able to present lines of reasoning with a capability to do drill-down to utilize the tracability aspect. Maybe presentations to some degree will be more in the form of a rendition with links back to text, data, graphs, images or whatever. We need to accept that in many cases it isn’t realistic to try to boil it down to summarized and instead be able to explore that data ourselves. Now, let us setup our mindset, software and meeting rooms to do just that!