Data Viz Thoughts .|: My UN Summit #VizFail and Alex Varlamov’s #VizWin
A Tableau-centric weekly blog about the viz making process, #datafam member interviews, #DataVizThoughts Viz of the Week + entertainment for introverts (consisting of a music morsel & a binge bite).
This week we will share thoughts on one of my vizzes + Alex Varlamov (@AlexanderVar7) ‘Evictions in San Francisco’ is our Viz of the Week (#VoTH)!
Before I begin the real blog/learning experience, the past few (Tableau) days have been otherworldly. I’m an aspie (this will likely be my only direct mention of this on the blog — other bloggers like Hunter Hansen cover this better than I ever would or could); I have ‘dull’ emotions compared to neurotypical people (unless it’s something of a specific topical focus that draws me in to the tipping point of obsession), but finding a community that I blend with and admire has tested that ‘dullness’.
Wednesday, my The Smiths music viz (inspired greatly by Nina Lindell) appeared near the bottom Data + Music viz gallery (a gallery that features the best music-related vizzes on Tableau Public across the world) & my Elvis viz and .gif developed in collaboration with Zach Bowders (covered on my 1st blog and check out Zach’s incredible gonzo blog) happened to be featured by Tableau on LinkedIn and Twitter just yesterday. I’ve rarely felt this elated in my life; I think because my narrow focus (Tableau) has a massive and diverse community with the same focus and great energy to prop each other up.
If it weren’t for Tableau Magic’s nudging, I would have never bothered joining the community, blogging or actively creating on Tableau Public.
At 44 years of age, I am able to connect my career and social niches. Not to be deluded… the Vizzes of the Week I feature here are ones I wish I could create — their passion, drive and talent are superior to mine; I guess that is prime reason I showcase them — these are the real viz artists that you should be looking to gravitate & the guy writing about them feels exceptionally fortunate to have the opportunity to engage & share.
Viz Thoughts on….
To my discredit… 1st World Vs. 3rd World Sustainability Goals
#MakeoverMondays is a fun-strating weekly challenge. The data can come from anywhere and usually needs an additional wipe-thru to prep for a viz. Nearly two weeks ago the ‘test’ was more meaningful than most because it fed from a UN Survey focusing on global goals — favorite vizzes from the week would be shared with the UN General Assembly (happened earlier this week).
The reference visualization was not helpful — sorry if you clicked the link because it (the Pennywise of vizzes) cannot be unseen. (1) 1st of all, staring at it was like staring into an eclipse with high-powered binoculars. I knew the only inspiration that could be drawn from it was a cautionary one. Next, I tried to figure out an angle — there were a lot of fields and directions one could go.
So how ‘bout that data?
The data was a bit mangled with no apparent data validation. Some people completing the survey were over 100 (topping out 140 years old) and five and under (bottoming at 1 years old). Removing those rows and others with similar (very likely) errors still left me with a huge data set and no direction. To get a better idea of the data and its context, I took the survey — this survey was slick but puzzling to complete — a confusing survey with no data validation is a disaster data credibility-wise.
When I’m frustrated but it something I set out to complete, my mind jumps to its happy place — memes. I know what you are thinking — Mico… you’re dealing with a serious topic with a reputable organization — how can your mind get meme-drunk? It did. I couldn’t get first world problems (2) out of my head. Yep, I had a surefire fail angle, but thought I could make it work.
Grabbing additional data and adding a field, I created a list of ‘developing’ vs. ‘non-developing’ countries (commonly exchanged with 3rd and 1st world) and make some reference to ‘1st World Problems’.
So many other thoughts were jumping in now…
1st world problems goals vs. 3rd world problems goals
Woot … there it is (Future Mico — sorry, there it ain’t)!
After that, competing bar charts with two interactive globes came easily. It wasn’t my one of my best vizzes, but it wasn’t an embarrassment. (3)
I posted to Tableau Public and shared it on Twitter and LinkedIn. The feedback was great & it seemed like people were digging the viz. I went to bed and woke up and then saw the bottom fall out; it turned out my 1st world vs. 3rd world angle offended people.
Mico (3rd person again… I know) was so ignorant and meme-drunk that he was stunned that it could be offensive.
Evelina Judeikytė (@Evelina_Jud) was not having it — she was courteous, but persistent and firm until the sleep cleared from my eyes and my meme hangover rendered me apoplectic — at that point, my silly brain was open to the premise problems. It made me want to learn because my thick skull makes it hard to perceive how this could be wrong, but something in me knew I fouled. Evelina directed me to Factfulness by Hans Rosling.
I read the book (on Kindle) and what struck me more than hurting feelings was that my facts were slanted by perception and completely misguided by not getting context or researching changes in global health/wealth patterns & applying so many misguided layers of sedimentary poop [emoji]— having a inaccurate binary outlook made me feel like a person passionately preaching the earth is flat or that Keanu Reeves is an immortal vampire (wait… that may be right).
The book also shared that there are many others far more educated and accredited than me with a similarly misguided/biased angle (driven by perception, generalizations and straight-line thought processes among other related miscalculations). The information Hans (I guess I think I know him like that) shared was also absorbing because his approach applied a very realistic/logical data practices and did not preach to me and ask me to change my being… just that we need to be guided by actual data and avoid tons of pitfalls that feed into dangerous continued misinformation.
Mico… so why is the viz still up?
I’m not taking the viz down — it’s a learning experience and maybe a vaccine that will prevent future meme-drunkenness. If I were to redo it, the entire concept would have to be rethought because there are so few countries so poor that the average person has little to no resources for medical treatment, basic nourishment or education. Many countries have a similar middle-class lifestyle that makes the comparison I made 100 years+ outdated. The global norm is that nations are continuing to advance to a progressively (more) comfortable lifestyle.
Viz of the Week
Alex Varlamov is a virtuoso of design and insight. There are very few that can ascend to his math-drawn visualization proficiency. His clever work captures Tableau’s bright future and boundless scope (of course he does use other tools to make the vizzes to soar to next-level mastery). He is also an active & helpful community member on social media (faults do not compute with Alex).
Adam Mico (AM): What inspired the 3-D line chart direction when working with the data (it’s absolutely perfect here because the lines resemble buildings so measuring evictions that way has a unique visual impact)?
Alex Varlamov (AV): I’ve seen a similar joy plots on different maps from the QGIS community and even on a book covers. This chart is not unique — it uses in geo visualizations and names ‘elevation profile lines’. The chart is not 3-D (4) — it is rather pseudo 3-D where there are no 3-D distortions and heights of points equal numbers of event (or evictions in our case).
Previously, I made a similar chart using world population data. The data in #MakeoverMonday Week 39 seemed absolutely acceptable for creating this type of chart because the data contains a lot of events (evictions) and thousands geo coordinates. Disclaimer: I can’t recommend using this type of chart in business visualizations due to pseudo-3-D effects because it may be misinterpreted. Although I hope to collect feedback from community and if this type of charts work better than heatmaps, for instance… so why not?
Another name for this chart is ‘Population lines’ for Population.
Here are some links:
https://spatial.ly/2014/08/population-lines/
https://spatial.ly/2017/04/population-lines-how-and-why-i-created-it/
https://reddit.com/r/MapPorn/comments/9lgbhq/population_lines_of_new_brunswick_showing/
https://hannes.enjoys.it/blog/2019/09/dynamic-elevation-profile-lines-as-qgis-geometry-generator/
AM: What may get lost is that you used ‘math’ to draw the boundary map and it’s not some image overlay. I do not recall seeing the ‘sketch’ design like that before. It’s remarkable when matched with the 3-D eviction lines. When did you determine to use the design as you did and how did it come together? Did you go into the viz to do a minimalist design and build around it or was it considered later?
AV: The formulas I use describes rounded latitude and longitude work on a sheet with a map (i.e. we can imagine the dots of the lines as latitude and longitude and put it on a map). If we need additional layers, we can make a base layer with the charts and put other transparent layers on the base layer overlaying points. Therefore, the other layers will have equals coordinates.
Concerning sketching… My first draft was colored and used 3-D map from Mapbox, but it looked so cluttered — that’s why I removed the base map layer and revealed only borders. The viz uses native Tableau map layers. You may have noticed the different layers do not have the same borders (area code boundaries & coastline, for example) — so we can achieve the ‘pencil drawing’ effect when a map has few lines for borders. I liked it and my thought was, ‘Is it possible to create viz as a pencil sketch using only native Tableau features?’. Making the viz, I noticed that area polygons (districts) overlap borders partly and we can see an interesting ‘eraser’ effects (eraser ‘remains’ somewhat erasing the ‘pencil’ lines). When I created ‘pencil style’ joy plots on a map, it looked boring and I tried to find ways to make color accents. I chose blue because it looked like a pen’s ink. This solution looked more attractive and for highlighting the effect I made different line density — thinner below and thicker on tops. The ‘pen effect’ was achieved and the line chart (joy plot) looks like somebody drew it by pen.
Music Morsel
On 9/28/1987, the Pixies dropped its 1st (mini) LP (Come on Pilgrim). All songs on this song are primo, but the album’s closer… Levitate Me best demonstrated their high-voltage swagger & signature loud-quiet-loud dynamic that helped change the sound of alternative rock music. This band’s influence is gigantic (5) as Radiohead, Nirvana, The Smashing Pumpkins + so many more have mentioned them as an inspiration.
Binge Bite
Netflix’s new short series, Decoding Bill Gates, is a must watch for anyone curious to see how Bill’s mind works. It deftly interweaves a biography and his current fight for global health (focusing on his partnership with Melinda). Bill also digs Factfulness.
Footnotes
(1) The clown antagonist from Stephen King’s It.
(2) First World Problems on knowyourmeme
(3) Mariona Banyeres (@MarionaBanyeres) was such a help. She knew that I was meme drunk, but also aware that I was determined to post something. Yes, she saw the 1st vs. 3rd world references, but with any meme drunk person… you give them a mug of coffee, take their keys and point them in the right direction — the vs. was just the Uber driver that came and took me home safely. She didn’t pick the driver — I did, but left my sense in the car.
(4) Thanks for the correction, Alex!
(5) Pixies fans will know what I did there. BTW, my favorite Pixies album is Bossanova.
Continue sharing #DataVizThoughts
