“january” progress

no noo

I’m finished with tagging entries up till 2010, although only entries up to 2009 have been uploaded on the site so far. The rest are on the Textedit file on my computer. I’ve been working with really unstable network connection and it’s driving me up the wall. I think this is one of the drawbacks of moving your work around and doing it on the go.

Also, I am quite frustrated with the amount of spam comments that are generated by my blog, even more so with the fact that Akismet offers the service for a ‘minimal fee’. I went with the free option anyway, but it does not make sense to pay to get rid of spam comments.

On the website, you will see a grid image. It’s meant to be a filler image at the moment. I will be writing a WordPress theme for this website, something close to a sketch I made a while ago. The entries are available for viewing (under the read more tag). I am still interested to have the metadata as the narrative. You can see that my tag cloud is quite specific.

The next step for my project is to plot in the number of entries per category/tag into Google sheets and creating a skeleton for a visualisation using the charts.

I am also paying attention to how I feel as a result on embarking on this project. It didn’t make me feel good to look at my own writing and experience in such close detail. I have the same feelings when I worked on the ‘dictionary’ project last semester. At that point of time, I felt mortified at some of the entries. I become more aware of myself, and that some of my flaws have been apparent for a long time, and I may not have looked into them or addressed them at all. But this are more personal reflections of my project that I think I’m not ready to share right now, without going through a long story about the roots of my angst, etc. I’ve been writing down some of these reflections as I do my work, and I hope to put them together as part of the conclusion for the project. All of this is a work in a progress, my personal self, the project… at the end of the day, I would like to be able to look at this project and know that I have made something good out of what I’m not proud of.


visualising january: part one



This week we begin preparing for our final project for Facts & Fictions. I’m taking this opportunity to apply my newfound skills to my current project by visualising the entries from January over the period of 2005 to 2015 (January x 10). The amount of data over this period is a good size to work with for the final project (which will culminate in a group show in another 14 days), and I think it is a perfect chance for me to try visualising a part of my FYP.

These are some of my process shots:blogtagging01

Filtering entries from January 2005: there are 17 items, of which I am going to individually tag and categorise them. In 2005, I was still on Blogger, and the blogging platform is very simple, and I don’t remember tags or categories existing on Blogger then. So it was quite a good thing that my entries are left un-tagged/categorised, now I can be very specific about how I want to label the entries for the sake of this project.


I am also doing tagcrowding again, with the help of Miriam Quick, who does research for information design. With her guidance, I learned how to use Tagcrowd in a more resourceful manner. She also taught me how I can use Google Sheets to my advantage, by showing me lots of cool stuff that can be done with Sheets. After running through my text in Tagcrowd, I went on to omit common words, and made a list of frequent words I use. This is different from individually tagging my entries, which I feel is something I would have to do manually if I really wanted to be specific about the topics that I wrote about, and I think there’s no shortcut to this part. Tagcrowding would be useful for highlighting linguistic details like: lingo, swearing, emotive words, and even names. This could be an interesting area to visualise on its own, so at the same time, I am also creating an additional dataset for that aspect.

blogtagging03 blogtagging04

I’ve installed WordPress on my own site again, for this January project. Currently, there’s nothing fancy there yet, but I think the taxonomy is taking shape and I am very excited about it. I’ve just finished importing entries from January 2005. You can take a look at it as I update it with more January entries, although I must warn you that some of the entries are very juvenile. Please bear with my 13 year old self. Haha.

Overall, this whole process of creating the metadata is far less agonising that I expected it to be. Before embarking on the January project, I read through all the January entries in the ten years, which left me in a very sombre and nostalgic mood, but all of that is gone when I go all technical about the work. That really gave me an idea to write a reflection piece after I have completed making these datasets for my archive. It would be interesting to include in my process book.

I’ll share more when I finish making the datasets for each year, and also my process on how I will visualise everything.


week one reflections

It’s been a really packed and intensive week! I have so much to share, and it’s amazing how much I’ve learned about data visualisation in just these few days alone. Here’s a quick recap of what I’ve learned over the week.

First, we have Stefanie Posavec, who was with us Monday to Thursday. On Monday, we gathered a dataset that represents the spirit of Berlin. I went to a local supermarket to collect some information on the products sold there. We learned how to make a physical visualisation out of the data we collected.

Photo 4-11-15, 8 02 32 PM

I made this pie chart out of salami, to represent the different kinds of wursts that can be found in a typical German supermarket.

Photo 5-11-15, 7 54 43 AM


I also collected another data set on the number of times I checked in with my family and friends over the week. I was having a cold this week, and the tissues were used as a form of visualisation.

Stefanie’s approach to data visualisation is very hands on: she does not do programming. Instead, she draws everything by hand. Her process is very well documented, and her slides contain many photos of her sketches from her notebook. She encourages us to try this approach on our last day with her, where she gave us a set of data on the weather patterns in a month.


This is my outcome.

Key things we learned with Stefanie: find a pattern in your data, select your focus, and always remember than form follows function. Set up rules. Create hierarchy within your focus helps to make your visualisation even clearer. Always create a scaffolding before inputting your graphical elements.  Think sequentially, incrementally.



On Thursday evening, we had an artist talk by Maral Pourkazemi. Maral is a very passionate designer. She has Iranian roots, which shapes most of her design work. She calls herself a design activist, where she uses her data vis skills to create powerful works to highlight the issues faced by Iran. This is one of the most interesting talks I’ve ever heard, as she shares the potential threats that surround her as a designer, because her work involves politics. Like Stefanie, her master thesis was the work that propel her to the world of data vis, and offered her many opportunities and led her to where she is today. For her masters, she made a work about censorship of Iranian internet.

We ended the week with Nicholas Felton, who shared with us the programming side of data visualisation. I was very interested to hear how he made all his Annual Reports, and everyone was curious about this whole effort of collecting data actively, consciously, for ten years and putting them together in his report. Each year, he challenges himself to make a better report than the last. His last few reports were made with the help of applications, a few that he developed by himself.

This weekend, Nicholas walked us through the program Processing. I’ve gone through a beginner’s tutorial the week before I came, so thankfully I was no stranger to this. Having just a bit of introduction and some background in coding helped me a lot in picking up Processing as I go along.

Screen Shot 2015-11-07 at 4.16.57 pmScreen Shot 2015-11-08 at 1.38.33 pm

The possibilities of what you can do with Processing is limitless. Learning the programming script is akin to learning another foreign language. One of the key things I’ve learn with Nicholas Felton on Processing is actually how to input data into the program. We made spreadsheets on Google Sheets, converted them into a .csv (comma separated values) file, which is readable on Processing, and then using coding, we are able to create visualisations out of it. The above screenshot with the lines is an exercise I am working on right now. The number of lines is generated by the values from my spreadsheet. This is all very sketchy at the moment, and I’m working on adding the graphical elements later on. This is another really important thing I’ve learned this week, and something I probably will not be able to figure out on my own.

So, week one in summary is learning how to create visualisations with the help of coding, or by hand, and I am really glad to be able to learn from Stefanie and Nicholas, who are very, very good at their work.

Something else I’ve picked up this week from them is actually how to present your work. Watching these designers speak about their work is so, so inspiring. They are very confident and clear about what they are trying to say, and this helps a lot particular at this point, when I am simultaneously working on my FYP prep. Everyday I have something new to add to my presentation.

This week is particularly intense, but I hope I will polish up my presentation and be good to go by Tuesday.


learning with stefanie posavec: day 2

thumb_IMG_8649_1024Today we learn how to visualise data. Some basic techniques are introduced, as well as some general rules of thumb as a guideline.thumb_IMG_8652_1024  thumb_IMG_8651_1024

Using statistical knowledge to work with design. Looking for an overarching conclusion may help make your work more meaningful. Patterns and rhythms in data can be translated visually – using various methods like analog, coding, etc.

Have a spreadsheet! Use Google sheets. It helps you to identify patterns.

Take notes on what you find: rate of change, hierarchal relationships, and so on. Get to know your data.

Select your focus. Form your message, find the highlight.

Assign visual variables to data (shape, tonal values, texture, orientation of a line, etc)

IMG_8662  IMG_8660 IMG_8659

Stefanie highlights the elements that makes for a strong data visualisation:

  • Good architecture + arrangement
  • Annotate appropriately: labels, legend, titles, axis, units, sources, attributes.
  • Don’ts: improper scaling, truncated axes, differences in perspectives (particularly in 3D visuals)
  • 3 to 8 groups or categories is good enough to communicate


This form of visualisation is the basis for her style of visualisation: the node link tree diagram. It’s good to research on data visualisation styles to give you a headstart.



Do a sketch first: how it works, then add graphical elements, and then annotations.

IMG_8665 IMG_8664 IMG_8663

Critiquing bad data visualisations: it should not be too confusing.

Lastly: some methods to organise data:

  • grouping information according to location (geo-spatial data)
  • alphabetical order
  • time
  • category (comparing categorical values)
  • hierarchy (relationship between entities)

Jacques Bertin’s visual variables

Gestalt laws of grouping.

That’s all for today! We are to work on an individual project for this program as well. I won’t be using my FYP for this due to the time constraint of the program. I’ll share more as I go along.