Saturday, July 31, 2010

Why I don’t like Outlook: Reason #122 – Folder Display

I am forced to use Outlook at work.  There are many reasons I don’t like this, including the fact that it doesn’t comply with normal email standards, and I cannot check my mail through my chosen client. 

Due to some silly rules, when off campus I cannot even use the Outlook client, but instead must use the web interface.  The web interface does not work with Firefox, of course. 

Take a quick look at the way it presents the folders:

hunt for the inbox

What do you see first?  There is a very unhelpful ‘popout’ effect going on with this list, even at the reduced size seen here.  The two least important folders are in bold with coloured numbers beside them: deleted and junk.  I have to hunt for the inbox in alphabetical order. 

Granted, my inbox is actually empty (yay!) so that’s why it isn’t bold, but it is so critical, it should be right at the top.  I hear often from colleagues about the wonderful usability labs at Microsoft.  They brag about the ability to recruit and test with ‘real people’ instead of the undergrads us in academia are often forced to use.  However, it doesn’t seem to help them solve basic usability issues like this.

Monday, June 7, 2010

Lambert Academic Publishing Continues to Spam

I was contacted today by a representative of Lambert Academic Publishing requesting that I change the title of my blog post "Academic Spam", in which I criticize their marketing tactics as consisting of sending poorly written spam enticing vanity publishing. That's right: Lambert Academic Publishing and VDM Verlag send poorly written academic spam.  Parse that and put it on the search results, Google.

In their latest message, Michael Davis, a Customer Service Executive, insists they are not a vanity publishing house.  The two central premises of his argument are that they don't charge the authors, and the publications are available on and in physical bookstores.  Anyone can post products on, and removing the pay aspect doesn't make it less vain to essentially self-publish.  Oh, and they offer a dedicated "Acquisition Editor" -- someone who apparently can't even write an email without English errors.

As I understand vanity publishing, it basically means you get a publisher to print up your thesis and sell it without proper editing or peer review.  This seems to be exactly Lambert's model but they don't charge you (in cash).  Instead, they charge you in terms of removing your rights to your own work (you can re-publish only up to 80% of your work).  In response to my original post, several people have argued that no harm is done in publishing with them -- it gives you something nice for your bookshelf and CV.  I suggest these authors find a local print shop and bindery and do it themselves.  Having a self-published on-demand book from LAP is likely to be a negative, not a positive, on an academic CV.  Experienced researchers will see right through this in a second.  It doesn't bring with it any weight in terms of peer review, it just shows your own vanity and your inability to make your thesis available freely online. 

LAP is essentially an on-demand publishing house.  I would like to know how many of the submitted theses actually go into the physical bookstores?  I would wager that the vast majority exist only as cover shots in listings and in print form on the bookshelf of the author and his or her proud parents.  They spam all of us in the hopes that someday someone with a gem of a thesis will fall for it and actually make them some money.

The letter I received today was obviously carefully written -- if they had taken similar care in their initial message, it may not have raised my suspicions.

Well, I decline to change the title of my post, and instead I will link to several others who also consider LAP's emails to be academic spam.  Steer clear of this company.

See also:

[Edit 2015 - my original post mentioned "Nigerian 419 scam" emails.  While phishing emails and email-based fraud is real, the association with Nigeria was racist and I sincerely apologize to all I offended.]

Monday, May 17, 2010

Tulsa Library: Bastion of Censorship Online

The Tulsa City library system has a shameful system of online censorship. 

I do not support internet censorship of any form, but I do accept that libraries may want to filter potentially visually offensive materials to protect other library patrons from inadvertent viewing. 

However, the Tulsa library wifi (which, by the way presents an invalid self-signed security certificate) takes censorship to an extreme level which I haven’t experienced anywhere else.

Examples of blocked sites:

  • Craigslist
  • Boing Boing
  • (!) (presumably to disallow one from changing the SafeSearch filter from ‘strict’)censorship

Libraries should be leaders of information provision, not home to the most restrictive forms of censorship.  I hope that this policy is a sign of the puritanical culture here, not a sign of things to come at libraries everywhere.

Monday, May 3, 2010

Visualizing the Gulf Oil Spill

In order to help emergency response and the public to understand the extent of the current Gulf of Mexico oil spill disaster, several visualization researchers and designers have published useful visualizations.

NASA’s Aqua satellite supplied this image of the Gulf Coast oil slick resulting from the explosion & sinking of the Deepwater Horizon platform.  It was captured with the “MODIS” (Moderate Resolution Imaging Spectroradiometer) instrument, and shows the oil as it follows the gulf currents across the region.

[click for original NASA posting and high resolution image]

NASA explains why a conventional photograph would not allow us to see the oil spill as clearly as this MODIS image:

“The oil slick may be particularly obvious because it is occurring in the sunglint area, where the mirror-like reflection of the Sun off the water gives the Gulf of Mexico a washed-out look. Oil slicks are notoriously difficult to spot in natural-color (photo-like) satellite imagery because a thin sheen of oil only slightly darkens the already dark blue background of the ocean. Under unique viewing conditions, oil slicks can become visible in photo-like images, but usually, radar imagery is needed to clearly see a spill from space.”

CNN presents an interactive timeline of the oil slick movements, as does the NY Times. The NY Times animation includes narrative of events directly on the map, as well as the use of dotted borders to encode uncertainty when showing predictions of spill extent. Both could be improved with visual encodings indicating sensitive geographical regions, such as bird sanctuaries, oyster beds, tourist beaches, etc.  However, it is impressive how quickly the media are able to provide useful visualizations to the public – the flexibility of the medium will allow these designs to be improved as the story develops.

CNN Animation: broad context, few narrations

NY Times Animation: closer context, annotations and prediction

Critique of Culture Colours

David McCandless has published a book of visualizations called "Information is Beautiful".  The visualization that is used on the cover of the book is a graphic of colour connotations across cultures:

[original post at]

First, let me say that this visualization, along with many in the book, is quite visually engaging.  The data is interesting, and it would make a beautiful poster.  That said, from an InfoVis point of view, I think there are issues with this design.

Most importantly, the rings are, due to the geometry, different sizes.  This means that the slices of each annulus get progressively smaller as you move to the centre.  So, the colours of western society are given more visual presence than those of South America. 

The location and dual nature of the legend means that to read this graphic a lot of back and forth referencing is needed. Also, it's difficult without at least faint gridlines to determine which ring each colour block is on.  This is especially true for blocks in isolation, such as the red block in radial 12.

I personally might have visualized this using a small multiples diagram -- one for each word, perhaps with a single vertical slice for each region.


It would not be as compact, but I think it would be clearer.  Of course, this is just a five minute idea and I'm sure it has lots of problems too.  His graphic is clearly designed as a poster, and as such, it works well and would look nicer than my idea hanging on the wall.

Tuesday, April 6, 2010

Visualization Hall of Fame

My former colleagues Irene Ros and Matt McKeon along with Yannick Assogba at IBM Research have released “Many Bills”, a follow up to IBM’s revolutionary social data visualization service “Many Eyes”. Many Bills draws our eyes on the right stuff, using colour appropriately to highlight thematic similarities and differences in the various bills and versions of bills considered by the U.S. House and Senate. Interactive drill down allows an analyst to see an overview and then focus on an area of interest. I like their use of horizontal scrolling – a technique entirely undervalued in visualization in my opinion.

This tool could be a lobbyist’s dream – or nightmare if the public starts to see what really goes into the laws that govern modern democracies.  Following on their social data expertise, Many Bills lets visitors create, save, and share collections of bills that are of interest.     Check it out for yourself at

The slick tour will get you started on analyzing this fascinating data for yourself, or check out my sample collection below on Canada and the environment. 

Monday, April 5, 2010

Visualization Hall of Shame April 2010

I’m settled at my new job, and I’m looking forward to getting back to this blog now and then.  I plan to feature critiques in the form of “Hall of Shame” and “Hall of Fame” examples of visualization each month.  I hope for more “Hall of Fame” examples, but it depends on what I come across over the month.  This instalment: poorly designed pie charts.

Pie charts generally are not very useful other than to give a vague idea of relative size. We just aren’t that good at reading angles precisely. Adjacency, colour, the absolute angle, and other factors all confound our ability to read pie charts. Add to that the frequent use of ‘visual bling’ such as 3D perspective which makes it virtually impossible to use the charts for anything other than eye candy. Even if pie charts are useful, the metaphor is that they are pieces of pie. Thus the pieces generally add up to one pie. This point is, unsurprisingly, lost on Fox News. One might expect better from journalists, whose job is ostensibly to inform the public on matters of importance. Even more disappointing is the second chart, from Business Objects, the `business intelligence’ arm of SAP – this is a company that makes visual interfaces as their core business, yet their own pie chart makes no sense at all. Shame!

foxpie bo-pie-chart-large

I don’t claim to even understand the data underlying these charts, but if I had to guess, I’d say perhaps the questions asked about each data item were independent.  That is, 70% of people asked “Do you back Palin?” replied “yes”, and 63% replied yes to the same question about Huckabee.  In this case, a bar chart would be more appropriate, to understand the relative levels of support.  To get data for a pie chart, the question would have to be “Pick one: Palin, Huckabee, Romney”.