If you read a lot about typography, you probably have read that the ideal measure (also called line length) that is recommended is between 45-75 characters (somewhere around 2-3 alphabets) including white space, with 66 as the ideal. According to Robert Bringhurst, an oft cited source of this information, beyond 75 is ok up to around 90, but that he discourages this use for deep reading tasks due to pacing and cognitive reasons. In doing my own research into this, I've run into a quandary however.

I've been researching how major news organizations set their paragraphs, and at least for online materials I found that after researching a good number of high profile organizations I found that the average measure between them was ~87 characters (53-119 swing). This, in some cases is well beyond the deep reading recommendation and it makes me wonder whether these major organizations have made small missteps in their typographic setting with regards to responsive layouts, or whether these ideals (words per line, flow of text, etc...) are slightly less rigid than what is commonly purported.

I'd like some clarification, or maybe just thoughts that aren't my own on this.

Polled Orgs: Al Jazeera, BBC, Boston Globe, CBS Sports, CNN, Chicago Tribune, Google News, Mashable, NY Times, New Yorker, NPR, Sports Illustrated, The Telegraph, The Times, The Verge, The Guardian, Time, USA Today, Washington Post, Wired, Yahoo News, PBS, Wall Street Journal, MSNBC, Bloomberg, Politico, Fox News, Breitbart, Huffington Post, Smashing Mag, A List Apart, Nielsen Norman, Stack Exchange, Typewolf

  • I like this question but you may have much better response asking at graphicdesign.stackexchange.com – tohster Mar 27 '15 at 0:35
  • Can you post a link to the actual data you've collected? Seems like the news sites that care about typography don't exceed the recommendations by much. – Adnan Khan Apr 26 '15 at 5:46

I'd like some clarification

There isn't any. There is no such thing as 'ideal' line lengths. There's a handful of small studies but they are all flawed in that they can't isolate line length very well as a single factor.

What makes type legible, readable, and just plain comfortable depends on a whole range of criteria:

  • the typeface design
  • leading (line height) and tracking (letterspacing)
  • how it's justified (and if it is, how hyphenation has been handled)
  • quality of printing
  • medium (screen vs. paper, etc.)
  • quality of said medium
  • color
  • contrast
  • length of content
  • and etc, etc, and so on and so forth ad nauseum

According to Robert Bringhurst

Typography is a niche field of art and design and also a highly opinionated niche field. There are a lot of opinions, many of them valid and based on real experience, but very few of them are in any way conclusive by any scientific definition.

They are perhaps a good rule of thumb barring no other context, but there's always other context and variables that need to be taken into consideration--including the good ol' "what looks good?"

I've been researching how major news organizations set their paragraphs

Well, there's two typical places news organizations set their paragraphs. One is on paper, the other on screen.

As for paper, remember that line lengths are there primarily to allow for a very flexible page layout grid so that ads can easily be fit onto the page. Ads pay the bills, and as such, they get top priority. By using narrow columns, you create a whole lot more flexibility in adjusting content to fit around the ads--not to mention that you can sell a lot more variety of ad sizes.

As for screens, again, there is no ideal line length in general, and, on screen, you really have no control over that anyways. You can suggest something, but the web is in the hands of the end-user. They have a multitude of devices, device orientations, browser settings, system settings, preferences, installed fonts, etc. As such, what ever the layout is, it needs to be flexible, and part of that flexibility is to allow line lengths to adjust as needed by any of those aforementioned variables.

or whether these ideals (words per line, flow of text, etc...) are slightly less rigid than what is commonly purported

In summary, yes. Or rather, they are majorly less rigid than what is commonly purported by a few folks.

(I find it common in UX that we latch on to data much more than we should as a lot of said data is vague or simply weak to begin with. This is especially true the more you cross over into issues of typography and graphic design.)

As far as some thoughts, that aren't your own Nikolas, I think that you're right about publications being less rigid on their line lengths. I edit a news/blog site and the line length on articles is greater than 90, but for this site, eyes on the page is just as important as having an article read. Whereas when I design sites for clients or we look at content that is unlikely to change, I try and fit inside the 2-3 alphabet rule. I think you'll find that some publications are just a little more loose with how they lay out their text.

If it's really important that you encourage your audience to read content, then stick with the 2-3 rule, and do all that you can to invite reading. If actually having someone read the content is less important, then you can afford to have line lengths outside this range.

Your Answer

 
discard

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.