Not actionable, but interesting? Definitely.
Flickr user jbum has an image of saturated and unsaturated samples of photos with varying tags. OK, in English he looked at pictures that were tagged different things. And then found the average color.
From his notes on the tag color graph photo:
"For each keyword in this picture, I downloaded all the matching thumbnails, up to a maximum of 1000 images, and averaged the colors, using a script.
The left side of each stripe is the resulting color.
The right side of each stripe is the same color, with the saturation cranked up.
Now, I must admit, when I began this experiment, I was really hoping for a more obvious result: money = green, spring = green, sex = pink, winter = white, death = black. Something like that… Right?
No such luck. Reality, as always, rudely intrudes."
From the comments, also worth looking at is The Color of Palo Alto art project.
For the architects in the house, you can create a sketch with Gliffy and then extrude it in Sketchup for a fly through. All for free. Times are a changing.
The default for gliffy appears as a network diagram, which I would still likely build in network notepad. But that doesn’t provide collaboration. One plus for network notepad is that most network administrators aren’t excited about the idea of posting a detailed diagram of their network on the web with any company. Architecture on the other hand, or flow charts of processes, seems like a good use of gliffy.
Via the Houstonist site.
Watched the video on bumptop from the link on 43 folders. Bumptop is a new desktop interface prototype. Many of the user interface metaphors are similar to a real (physical) desktop; moving files around, stacking, shuffling things. The critique on 43 folders, linked above, addresses several of the limitations.
On the PLUS side, the "zip" motion of circling and drawing a line through the middle is innovative. I can see using that interface in favor of the CTL-CLICK-CLICK-CLICK method I currently use. Circle, zip through the middle, post to flickr. Easy.
In the diagram note the large cluster on the top left. That is called a spanning tree diagram. All it would take to add more of those lonely trees to the primary tree is one connection.
A quick google search leads me to believe this is the original study from Jefferson High School.
Walking from Sharper Image to zGallerie at the mall made me connect the Phillips Digital Picture frame with a large white frame with a small 8×5 phto area at zGallerie. Given visualization is one of our focus points this year, it quickly turned into a project to combine the two elements into a Primal Brand story.
This was a quick project, no plans, just sawdust and scraps from the get-go. Pretty much what happens when a geek has tools.
The end result is pictured at left. Of course the photos rotate. The active photo of the word "ELVIS" was taken at the Elvis car museum in Memphis last year.
The build process is illustrated in pictures as follows. I would like to issue a disclaimer on the drill visible in some of the photos. It is NOT my primarly drill. Like all good geeks I have a Dewalt professional. But of course I leave it at work, so what you see is my antique hand me down Black and Decker with the old-world drill bit chuck. The death of an american brand….
The main point is onward to the project pictures!
This visualization from Yahoo research is very interesting. I attended a talk at eTech where the speaker diss’ed the famous Minard’s Napoleon’s advance graph saying "but it didn’t help Napoleon make a single decision." The audience of course vehemently protested, echoing Tufte in a manner like Trekkies quote Spoc, "but it wasn’t designed for that! It was a antiwar protest poster!"
Reading the comments on the digg story, the best part is the link to the paper explaining the technique used. Data mining such a large time sensitive data set is not easy. I like the fact they explained how they did it.
We recently closed a client in the Hispanic advertising space so I was reading the AdAge Hispanic Fact Pack. This, crossed with recently going through the initial tutorial in Google Sketchup made me want to play with the pie chart as a 3D infographic for better visualization.
I did not do that. By no stretch does this graphic do a better job of conveying the information than the flat pie chart. This is just a proof of concept.
But, it was a fun journey and I thought it might be helpful to explain the process of converting a pie chart into a 3D visual using Sketchup. Here are two proofs of concept.
After the jump you can see the step-by-step process used and download the source files to continue editing it. Feel free to extend and improve.
Of course this also begs the question of if we WANT to work in 3D space. Play? Absofreakinlootly, but work? I really don’t know as I haven’t had the opportunity yet and only if it can be configured to expand horizontal adjacency.
In between witty binary to text email conversations today (it IS both Friday and Cinco de Mayo), I was sent a link to Fixing Excel charts”¦Or, why cast stones when you can pick up a hammer. This is just brilliant.
The short version is you get an option in Excel that says "clean chart" and your Excel charts look better. Here is a visual of what the most basic Excel bar chart looks like before and after running the "clean" command.
This may seem like a slight difference, but it saves a HUGE amount of time cleaning up difficult visuals. My hat is off to the brilliant folks behind this excel charts visualization plugin! Visit that last link for the downloads and installation instructions. Really, wow.
And before I forget, 01010100 01101000 01100001 01101110 01101011 01110011 00100000 01001011 01100001 01110100 01111001 01100001 00100001
Via metblogs on traffic by katya, this visualization of Houston traffic on the omninerd site is awesome. Mostly because for this specific data, the traffic hell that is Houston, a boxplot is the way to go. It shows quickly and easily that Monday and Wednesday suck. And that perhaps you can leave a few minutes later on Friday and still get to the office 15 minutes early.
Can we get a boxplot on Houston Transtar?
Let your eyes guide you. You look at the data visually. So you look at the bar chart. You assume 88% of society is OK with adultery. But that is not what the date says!
Then, and only then, in shock, do you notice the text to the right that qualifies the data with "percent saying this behavior is morally wrong". That is sensationalism. Visual sensationalism that undermines trust. This is a bad visual because it is effective while at the same time just wrong. Be honest damnit, especially with visuals.
Disraeli? Oh, he is the guy who said:
‘Lies, damn lies, and statistics’ – Benjamin Disraeli
What he meant to say was:
‘Lies, damn lies, statistics and then we have bad visualization’ – Benjamin Disraeli (cough – edited – cough)
I wish I could give credit to the correct person for this particular … visualization. I can definitely give credit to Jennifer for forwarding it to me by email. Thanks J. I think.
Whenever I meet someone for the first time they say "did you know there is an airport in Amsterdam by the same name?" – me: "yes" me: "and it is spelled a bit different" – them: "no, its the same" – me: "ok, I guess they changed it" – or something like that. I wonder if people named smith get "Oh, I have a cousin with that SAME NAME! Amazing!" – or something like that.
This is a funny small element of a urinal that has a bottom line result (sorry for the pun) that increases profit by reducing cleaning costs. Humans are funny. Visuals matter, sometimes for humorous reasons.
The text in the Schiphol urinal image reads:
In Amsterdam, the tile under Schiphol’s urinals would pass inspection in an operating room. But nobody notices. What everybody does notice is that each urinal has a fly in it.
Look harder and the fly turns into the black outline of a fly, etched into the porcelain. It improves the aim. If a man sees a fly, he aims at it. Fly-in-urinal research found that etchings reduce spillage by 80%. It gives a guy something to think about. That’s the perfect example of process control.
If anyone knows the origin, please email me or comment below?
Sparklines, like this one
, were created by Edward Tufte. Sparklines have gone live on all Tendenci association web sites. The image below is a crop from an articles management report, but the actual sparkline is visible to all visitors to any site. The idea is to accept the fact that the visitor is intelligent and would also like to see the performance of an article among other readers.
One trend visible on other sites with much content locked down for members only is a huge reduction in views of restricted content. This is intuitively obvious.The sparklines help make the case by hitting you over the head with obvious visualization of what the users are doing.
People do not like to be hassled. They don’t want to go to your site, login and search when they can go to google and search across multiple sites faster. The result is locking down your content causes it to go unseen for the most part. Don’t blame the sparklines – they are just telling you the facts.
Everything is relative. Julian Beever sidewalk art at its best. Thanks for the amazing visual forward TJ!
"Working with layers, allows me to build up subtleties of color that psychologically impact the viewer. Through the use of color, I hope to jostle the viewer’s memory of a particular place, reminding them of a moment they may have had." – Cheryl Tamborello, Houston Artist (emphasis added by me)
Sort of an attempt at abstract random mnemonics (memory aiding devices) for the viewer. Performance art in static form through visuals. The same thing that makes visiting the Rothko Chapel an obligation if you are in the Houston area (pictures never do Rothko justice as I am sure is the case with Tamborello as well).
While the three main learning styles are visual, auditory and kinaesthetic it has been my experience that visual is dominant in almost every individual I have worked with. This might be self selection based on industry but I doubt it. You can test learning styles and determine that someone is an auditory learner, but when asked "what car did you drive in high school?" they will picture it first before reliving the engine sound or some other related sound in their mind.
A visual mashup of airplane tickets. This is why I am so excited about great visualization. Huge amounts of data, HUGE AMOUNTS, are easily understandable with the right visual. Not only is your visual memory unlimited, but millions of data points are easily understandable to a 12 year old with the right visual. Patterns are apparent, lumpy lines are best according to Tufte, at a glance.
A good example of this was highlighted on TechCrunch today by Nik Cubrilovic in Flyspy Brings The New Web To Airline Ticketing. First the good news from the post:
Flyspy reverse engineers some of the mystique associated with the airline industry and makes it extremely transparent.
The search result, which returns very quickly, will present me with a graph of flight prices over the next 30 days so that I can quickly look at which days are the cheapest to fly. To book a flight I just click on the point in the graph. Simple.
Overall a very cool service that epitomizes what the new web is about. Flyspy should go public in a few months ““ you will hear more about it here and probably feel the distruption.
Now for the bad news. You can’t trademark or patent a visual as far as I know. And this is fundamentally a line graph. So there must be some other differentiation. There are definitely a few investment problems with mash ups. So I love it, but I wouldn’t invest in it without some form of protection. I definitely wish them the best.
The Dumpster is an interactive online visualization that attempts to depict a slice through the romantic lives of American teenagers. … The project’s graphical tools reveal the astonishing similarities, unique differences, and underlying patterns of these failed relationships, providing both peculiarly analytic and sympathetically intimate perspectives onto the diversity of global romantic pain.
From a visualization and ethnography perspective it is worthy of recording. It would be like if Flash had never experienced Joshua Davis as a leader. Yet like much of Joshua’s work, there is something missing. It is cool, but what did we LEARN? Is it in any way predictive? Interfaces are great, but can we load up our friendly neighborhood neural net and get cranking away on what the opportunties are for the team next week!?!