Ben Halpern

for The DEV Team

Posted on Jan 14, 2019

Changelog: Reading Time Now Displayed

#changelog #meta

The approximate reading time for a post on DEV is now displayed on feed pages as well as at the top of the post.

Feed

Post Page

It will likely also be added in places like liquid tags in the future, and anywhere else it could be contextually helpful.

Happy coding!

Top comments (26)

ItsASine (Kayla) • Jan 15 '19

Doing a quick regex on my Reading List (670+ right now), 40 were caught with a quick [0-24-9] min read while 500 were caught with 3 min read. This is fallible for multidigit things like 13 min read, but apparently I don't tend to save long form posts for later. 😅

Thanks for the extra data points!

Ben Halpern • Jan 15 '19

We actually still have some older posts which have not yet been indexed and are defaulting to 3 mins. This should be rectified soon. It’s a known issue but I figured people wouldn’t notice this before it was ironed out. Leave it to devs to quickly uncover the problems.

ItsASine (Kayla) • Jan 15 '19

Leave it to devs to quickly uncover the problems.

QA is my day job ;)

Gareth Bradley • Jan 15 '19

I love that someone (Kayla) bothered to investigate this. Awesome.

ItsASine (Kayla) • Jan 15 '19

I actually typed up a comment originally like "Uhhhh this feels buggy" but given how much I've read on this site, having a ton of stuff in the 3-minute range wasn't that unheard of. Occasional long-form posts, sure, and occasional discussion questions without context, but for the most part, people tend to succulently say their piece.

Meghan (she/her) • Jan 14 '19

Cool! How is it calculated?

Ben Halpern • Jan 14 '19

  def calculate_reading_time
    word_count = @content.split(/\W+/).count
    (word_count / 275.0).ceil
  end

Contributions welcome 😄

Vinicius Kiatkoski Neves • Jan 15 '19

Ben, just wondering, why 275 WPM?

This question came up to my mind and I decided to read a bit about it and found the following reference which recomends 200 WPM: marketingland.com/estimated-readin...

There is also this from Wikipedia where there is a nice graph with several different studies correlating WPM / age.

As I mentioned I'm just curious about the magic number (where you've found it).

ItsASine (Kayla) • Jan 15 '19

From your Wikipedia link, this caught my eye:

reading for memorization (fewer than 100 words per minute [wpm]); reading for learning (100–200 wpm); reading for comprehension (200–400 wpm); and skimming (400–700 wpm)

I'd think most here are reading for learning, to an extent. 200 WPM sounds like it'd be a good middle between learning and comprehending, though I could just be slow since I tend to read at work while I'm watching tests run.

Ben Halpern • Jan 15 '19

Hmm all good food for thought. We settled on this a while back and I sort of forget the exact reasoning.

We’ll make this smarter over time.

Vinicius Kiatkoski Neves • Jan 15 '19

Just a tip (don’t wanna bother you with that) but usually when I get those magic numbers on my code I try to add comments linking to it. It might help future contributors to understand and think twice before updating it to another magic number hehehe
As I mentioned I was just curious about it!
Nevertheless I think it is a great feature and it might help engagement ✌️

Miguel Piedrafita • Jan 15 '19

Haven't worked with Ruby a lot, but would you be open to a PR that takes images into account?

Avalander • Jan 15 '19

Althoug this is a great metric, I personally would find more useful to see the word count directly. Would you consider something like this?

8 min read (~2.4k words)

Steve Layton • Jan 15 '19

Ohh fancy!

Anton • Jan 15 '19

When I saw that feature I didn't like it because:
1) it's clogging space. (less is more kinda thing).
2) what if the article is technical? I'm sure to read it more thoroughly so the time is so relative that the purpose of the feature is counter productive. It doesn't tell me any REAL information.