Show HN: Tufte's discrete sparklines using D3.js
dataviztalk.blogspot.com
Show HN: Tufte's discrete sparklines using D3.js
1–9 of 9 posts
Re: Show HN: Tufte's discrete sparklines using D3.js
#2Curious what the advantage is (in general) of sparklines over, say, a line of circles with filled in = win and empty = loss? you could use a red fill color for streaks of 4 or more. It seems to me a solution like that would have less clutter and better communicate the data...
Re: Show HN: Tufte's discrete sparklines using D3.js
#3What does the partial horizontal line indicate in Tufte's example?
Re: Show HN: Tufte's discrete sparklines using D3.js
#4What does the partial horizontal line indicate in Tufte's example?
Looking at the Bosox 2009 schedule, it looks like the horizontal line indicates home games.
Re: Show HN: Tufte's discrete sparklines using D3.js
#5Curious what the advantage is (in general) of sparklines over, say, a line of circles with filled in = win and empty = loss? you could use a red fill color for streaks of 4 or more. It seems to me a solution like that would have less clutter and better communicate the data...
A single line of circles allows for less information density. If you force it to have the same information density then you lose a lot of legibility. Secondly, it's possibly harder to see where outliers are. I find this especially true when the vast majority of the dataset is 0, and you're looking for the outlying 1s.
See this example I mocked up, where each line has the same number of data points: http://i.imgur.com/xRoY41c.png
Re: Show HN: Tufte's discrete sparklines using D3.js
#6What does the partial horizontal line indicate in Tufte's example?
Looking at the Bosox 2009 schedule, it looks like the horizontal line indicates home games.
Indeed. 2nd paragraph on the linked page:
"This example is from Tufte's site, and was made in collaboration with Adam Schwartz. The presence of a horizontal line represents a home game."
Re: Show HN: Tufte's discrete sparklines using D3.js
#7Curious what the advantage is (in general) of sparklines over, say, a line of circles with filled in = win and empty = loss? you could use a red fill color for streaks of 4 or more. It seems to me a solution like that would have less clutter and better communicate the data...
A single line of circles allows for less information density. If you force it to have the same information density then you lose a lot of legibility. Secondly, it's possibly harder to see where outliers are. I find this especially true when the vast majority of the dataset is 0, and you're looking for the outlying 1s. See this example I mocked up, where each line has the same number of data points: http://i.imgur.com…
Thanks for taking the time to do that.
Re: Show HN: Tufte's discrete sparklines using D3.js
#8Curious what the advantage is (in general) of sparklines over, say, a line of circles with filled in = win and empty = loss? you could use a red fill color for streaks of 4 or more. It seems to me a solution like that would have less clutter and better communicate the data...
A single line of circles allows for less information density. If you force it to have the same information density then you lose a lot of legibility. Secondly, it's possibly harder to see where outliers are. I find this especially true when the vast majority of the dataset is 0, and you're looking for the outlying 1s. See this example I mocked up, where each line has the same number of data points: http://i.imgur.com…
Thanks, that example illustrates the information density point very well!
Re: Show HN: Tufte's discrete sparklines using D3.js
#9Would have been better had you used SVG.