Hacker News new | ask | show | jobs
by hinnisdael 696 days ago
Great advice: describing things in order of importance.

Most people intuitively describe images from foreground to background or left to right, a bit like they are mentally completing a checklist of all the things to describe. As correctly noted by the author, describing by importance first has the added benefit of allowing screen reader users to skip irrelevant/uninteresting images early.

Compare:

Torn-up painting in a gallery, observers standing in front of the work.

vs.

Gallery interior, people standing in front of a painting with visible damage.

4 comments

> Most people intuitively describe images from foreground to background or left to right

I've heard there are cultural biases to this ordering. Some cultures tend to describe the background or scene first. The example I read about identified Japan as an "outside in".

I've been encouraging my kid to use "outside in" or "context first" in their descriptions with me, mainly because they suck at giving context. I doubt I'd have known about it if I hadn't read that about Japan.

Would love to hear from a Japanese person on this.

I can't speak from a Japanese perspective, but there's definitely a specific-to-general or general-to-specific nature to languages.

"If you can get it in the next 60 seconds there's a million dollars in the jar in the back right of the second shelf of refrigerator." My wife is Chinese, she works from a general-to-specific worldview and I would be surprised if she could follow those directions in the time limit.

Flip it and say "If you can get to it in the next 60 seconds there's a million dollars in the refrigerator, on the second shelf, in the back right, in a jar." and she would have no problem with it.

We focus on the specific, build up the image of it's environment and then paste it into the world. She refines the world so all the references are left hanging until the whole sentence is complete and that's probably enough to confuse her. We just find her approach to feel odd but we can still follow it much better than she can follow ours.

There's a reason pretty the majority of electronic addressing uses that second format***. It's just easier to follow, and in fact you can start executing before the sentence is finished - as well as discard any specifics you don't care about without having to scan the entirety first.

Following the second instructions works like this: Go to refrigerator, go to second shelf, go to back right, go to jar.

Meanwhile following the first set of instructions requires this: remember jar, remember back right, remember second shelf, go to refrigerator, recall second shelf, recall back right, recall jar.

It is objectively a worse way to give instructions, because you are giving them in reverse order.

If you had to code some software following instructions and your interface was push(instruction)/end(), your first sentence would require more code and more working memory just to store instructions until you've given the last one, while for your second example they could just be executed right away.

*** Funnily enough it still tends to get flipped the closer we get to natural language. URL paths, domains, IP-addresses, etc, are all written the proper way, except if you add usernames into the mix. Suddenly we turn natural language "at" into "@" and prepend them.

For most purposes the second format is superior for the very reasons you state. In a certain world it's always going to be superior.

However, it means that concepts which are not attached to a worldview become harder to work with. Think of making subassemblies of jigsaw puzzle pieces. You can see that pieces are related and fit together without knowing what you're actually looking at.

I think of this in cinematic terms. The first sentence is going to start with a shot of the painting, then it will either cut, zoom or dolly out to reveal the crowd. Whereas the second shot starts with the wide angle of the gallery and then does the opposite. Each has a slightly different effect on the scene and the audience.
> As correctly noted by the author, describing by importance first has the added benefit of allowing screen reader users to skip irrelevant/uninteresting images

Good rule of thumb, but also very context dependent. Consider the following example of a pattern used extensively by Douglas Adams, and a little by Terry Pratchett, in service of humour:

    I made my escape by swiftly sliding down the laundry chute, my fall
    being gently broken by the Wednesday rota for large duvets. A perfect
    plan to avoid spraining the other ankle.

    Too bad today was Thursday.

For a dungeon crawler:

    You breathlessly take in the wonder of the large underground
    cavern; the twinkling of glow-worms high above resembling
    the night sky, the luminous ore-lines tracing sinuous veins
    away into the horizon, the mountains in the distance, at once
    both masking and highlighting just how large the cavern is.

    By comparison, the rapidly approaching 300-foot high fire-
    breathing Dragon intent on devouring your party fatally appears
    unimpressively small.

I can keep this up all day, actually.

The point being, punchlines have to go last.

Yes I actually liked the dragon being last. It created the sense of surprise over a calm unimpressive room. First the area was described as much as it would be if it was in real life and you were entering the room and scanning it with your eyes until your eyes finally go to this dangerous thing you didn’t see at first.

I think it’s better for narrative but for communication for sure the most important things come first, and sometimes last only if you want them to continue to have it in their mind.

The first sentence leads me to imagine a torn up painting and a group of people clustered around it.

The second sentences leads me to imagine a large gallery space with high ceilings with a smattering of people in front of one of the paintings.

Both ways have their pros and cons. Describing the space first lets the reader paint a setting for the eventual object of interest.

That's true, but consider the context; this isn't a novel, it's functional text for someone who is probably trying to accomplish a task. A user in a hurry might skip part way through the second description and be misled to thinking the photo was just a normal picture of a gallery.
Absolutely true in the context of the article ie alt text. I was speaking more universally.
What task are they trying to accomplish with a picture of a torn up painting?
Who knows? Insurance adjuster, private investigator, art historian, security consultant…
I think GP is a great comment with a poor example, because I agree that the resulting images in my mind are quite different, but they don't inherently have to be due to the order things are described in.