Talk:Chartjunk

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Notes on a sentence I added[edit]

"Although Tufte's book is widely regarded as a classic" ... published 1983, continuously in print, as of 2004 Amazon sales rank is 1621, Boston Globe called it "a visual Strunk and White," Whole Earth Catalog called it "THE visual style book," and the American Statistical Association said it was "The most important contribution so far to the study of the graph"--according to [1] but I see no reason to doubt either the accuracy of the quotes nor to suppose they are distorted by being taken out of context.

It's a classic. I first learned of it from the Whole Earth Catalog... before they were calling it THE visual style book, I believe. The American Statistical Association is a big enough gun to cite. And, no, I don't think it's necessary to verify the quotation. Wait for someone to seriously dispute it! Dpbsmith (talk) 14:50, 10 December 2005 (UTC)

"...as of 2004 the term is mainly associated with Tufte and has not achieved general currency." Only 904 Google hits on "chartjunk", not very many. Only 337 hits on "chartjunk -tufte" meaning that about 2/3 of them are in connection with Tufte. Not in AHD4. Not in common use to my own knowledge. And not surprising, as it's a rather awkward coinage (try to say it aloud—it's almost a tongue-twister). [[User:Dpbsmith|Dpbsmith (talk)]] 00:20, 11 Sep 2004 (UTC)

In December 2005 I'm getting 13,000 google hits for chartjunk, but still only 384 for chartjunk -tufte. I'd say that the prominence of the concept has become widespread, but it hasn't entered the general lexicon, as the vast majority of uses still contain a direct reference to the term's originator. To contrast - there are 23,800,000 hits for "cyberspace", with 22,500,000 for cyberspace -gibson (who popularised the term) and 23,600,000 for cyberspace -vinge (who coined it). Viveka 13:48, 10 December 2005 (UTC)

Merge with Tufte article[edit]

If chartjunk is indeed still associated with Tufte, ought we merge this article with the (surprisingly short) main Tufte article? Remes 20:05, 6 Apr 2005 (UTC)

I think there's enough more detail that could be provided about definitions and kinds of chartjunk, and arguments for and against it, that it deserves a separate article -- and some of the text here would seem out-of-place in Tufte's article. I agree that the Tufte article could use expansion, though; and moving some more detail on chartjunk there would be reasonable. -- Rbellin|Talk 20:11, 6 Apr 2005 (UTC)
I don't think it matters a great deal. I thought, and IMHO Viveka has confirmed, that it is still not terribly common and still is associated with Tufte. A short summary of the some of the basic principles he articulated in The Visual Display of Quantitative Information would not be out of place there (data ink, etc.) and "Chartjunk" could be a subsection or something. But no need to be precipitous. Dpbsmith (talk) 14:50, 10 December 2005 (UTC)

This should not be merged with the Tufte article. He coined the word for an existing phenomenon -- it is not his word and it describes something that existed long before Tufte gave it a name.

Better example[edit]

The picture is really an awful example -- it's just an awful chart. Surely someone has a better PD example, maybe from the US Government. Many of the examples in Tufte's book are also PD, so we could just use one of those.

The legend is also misleading--what does it matter what the file size is? It's only indirectly related to the amount of chartjunk in the picture. The real chartjunk is the visual artifact (the unnecessary fill patterns)
Done. Schutz 08:48, 24 July 2007 (UTC)
Done what? The current example is still terrible. --jacobolus (t) 20:19, 7 February 2008 (UTC)
Schutz changed the legend in response to the comment directly above his. I agree there are plenty of better chartjunk examples out there, go for it. --CliffC (talk) 21:41, 7 February 2008 (UTC)
I'll make a better example as soon as I can. Schutz (talk) 08:43, 8 February 2008 (UTC)

How about[edit]

Gratuitous backgrounds?

76.117.247.55 (talk) 01:27, 13 March 2009 (UTC)

Or alternately[edit]

Illegitimate 3D effects

76.117.247.55 (talk) 02:08, 20 March 2009 (UTC)


Who lifted from whom?[edit]

Either the initial three paragraphs of this article were lifted from http://www.synergy.com/chartjunk.htm or vice versa. (However, the quotes and references are ours, at least.) It depends on whether or not Synergy's KaleidaGraph page predates March 17, 2007, which is when the paragraphs in question more or less received their present wording.

The Wayback Machine did not index the page, so I can't tell. I guess I'll leave the paragraphs in for now.

--Ardonik.talk()* 06:15, 12 October 2009 (UTC)

Project Chartjunk[edit]

We need a Wikipedia project for this topic. Chartjunk everywhere. — Preceding unsigned comment added by Fluous (talkcontribs) 19:15, 14 March 2013 (UTC)

Chartjunk and Memory[edit]

Great article, but including it here is improper. Chartjunk is not mentioned anywhere in the article.

On the psychological side, it is related to the number of objects an average human can hold in working memory, which is 7 ± 2, according to G.A. Miller (1956).[1] The span of immediate memory and span of absolute judgment imposes severe limitations on the amount of information that humans are able to receive, process and remember;[1] it is necessary to select and focus on key information."

Fluous (talk) 06:31, 16 March 2013 (UTC)

References

Proposed merge with Misleading graph[edit]

This looks like a content fork that could easily be covered by Chartjunk in a subsection. Also, I am not convinced that "misleading graph" is a real search term, whereas chartjunk definitely is. We don't need to have two articles covering the same basic topic. KDS4444 (talk) 21:26, 11 September 2017 (UTC)

  • Oppose - the content is clearly different. "Misleading graph" is about displaying data in a misleading manner; "chartjunk" is about displaying extraneous elements in a distracting manner. --Jtle515 (talk) 19:16, 24 October 2017 (UTC)
  • Oppose - as mentioned by prior user, this article is about displaying data in a way that distorts the message. "chartjunk" refers to adding additional elements that are unnecessary or distracting. Those elements may or many not mislead or distort the viewer. Misanthropic789 (talk) 20:15, 8 November 2017 (UTC)
  • Oppose - Chartjunk adds "ink" without adding data, in Tufte's terminology. I have heard some forms of it described as "dancing bologna" by one freelance technical writer. Misleading graphs may involve chartjunk (or they may avoid it, to present a facade of gravitas) but mainly rely on other techniques; they are more likely to truncate real data or obfuscate the scale of it than to add extraneous irrelevancies. Just plain Bill (talk) 00:49, 8 December 2017 (UTC)
  • Oppose The two concepts are quite different: "chartjunk" involves the addition of visually distracting elements which are not misleading, but which distract the viewer, so that the message of the graph is not clear; a "misleading graph" is a graph which likely to mislead the viewer into making an incorrect interpretation of the information displayed MargaretRDonald (talk) 19:21, 28 May 2018 (UTC)

Paragraph 3 needs some examples[edit]

"Another kind of chartjunk skews the depiction and makes it difficult to understand the real data being displayed. Examples of this type include items depicted out of scale to one another, noisy backgrounds making comparison between elements difficult in a chart or graph, and 3-D simulations in line and bar charts."
1. items depicted out of scale (example needed +- deletion: this should probably go to misleading graphs and be deleted here.
2. frequency charts as 3-D, popular in Excel and quite meaningless
MargaretRDonald (talk) 20:34, 28 May 2018 (UTC)