Thread by @petermckeever

Wrote a longish reply to a message about viz but thought I’d share here for further input.. below are some of my thoughts but please add your own to help people looking to improve their graphics

Question: do you know of any good courses/material or just have any advice on how to improve the actual design process of data vis? Not in terms of the actual coding … but how to select the right spacing, colour schemes, fonts, etc…”

These kind of questions are difficult for me to answer. I’m very critical of my own work and I find being self-taught in coding and graphics means I’m probably the last person to give anyone advice on this.

But I do know I can create nice viz etc… so this is probably not useful advice but it’s what I have done:

interactive or static graphics that catch your eye: save them or the link to them. Have an initial look and write down what you think makes it work, what catches your eye, what have they done to make it look more in their style guide or where have they done something different.

What font are they using? Are they using multiple fonts? How do they make that work?

Every thought that comes to you about why you were drawn to or pushed away from a graphic: write it down. Sometimes it’s our own biases that lead us to disengage and it’s important to recognise that too (personal growth and shit)

Then? Forget about it. Let that sit with you for a couple of weeks (do the same with other graphics in that time), then come back to that graphic and do the same. Analyse the graphic, write what does and doesn’t work and why you might think that way…

Usually well designed and thought out graphics will evoke the same response but the time away will give you a different perspective.

I find this personally with infographics from the SCM,FT, some NYT. It’s not that I haven’t missed something in the first reading, great graphics convey meaning and help tell or tell the story, but there is always a sense of discovery on rereading..

Then compare your notes. Is there anything that remains true? Anything that has changed from the first to second reading? Then take those insights you developed and look to put some of that into your own work.

I know that sounds like a cop out but it’s like writing. You only improve by learning how to really read and by writing

You can’t write an original baby shoes never worn but you can as writers have take a tonne of inspiration from Hemingway .

Books and people that generally deal with data viz I would recommend: Andy Kirk Data Visualisation, Albert Cairo The Truthful Art, Jonathan Schwabish Better Data Visualisation, Lisa Rost’s post. They are my peeps from the top of my head anyway.

Ends I guess.

Caveat all of this by saying I’m not an academic and I’ve not had an apprenticeship to learn this craft so I’m very much a DIY Pete and there are much better placed people to give much better advice but those are my thoughts

Also John Burn-Murdoch, Russell Goldenberg, Amelia Wattenberger, Shirley Wu, Nadieh Bremer, Mike Hall all do awesome awesome wonderful things and are the main people I look to when I study static and interactive viz

SCMP*