Live data from Hacker News

Stop using icons in data tables

medium.com

51–60 of 66 posts

Re: Stop using icons in data tables

#52
No, this is wrong.

> When we pack high-density information into a data table or a complex dashboard we are increasing the visual entropy of the entire system. Forcing the brain to decode intricate, non-universal shapes in a tiny 16-pixel footprint, creates a “cognitive tax” that users pay en masse every time they scan the table.

This is not how the brain processes icons. Icons have their meanings embedded through related patterns. If you rely on common icons such as x, checkmark, plus, shopping cart, arrows, three dots, etc. the user already associates certain meaning/behaviors with these icons (and will be surprised if they act differently in your application). Introducing new icons is more challenging, because you have to convey their meaning to the user. At no point does the user have to stare at an icon and try to "parse" it pixel-by-pixel.

In all of OP's provided examples, the icons are supported by both labels and colors, which removes all ambiguity. You can argue that the icons are pointless in this case, but they certainly do not increase any cognitive load. Is it harder to identify a tiger than a lion, because his fur is covered in stripes?

Re: Stop using icons in data tables

#53
post #12

The title is misleading. The message is really: Stop using unlabeled icons in data tables. It says, "Norman Nielson argues that text + icon has the highest cognitive recall and lowest error rate" Here's what the Nielsen Norman Group says about Icon Usability: https://www.nngroup.com/articles/icon-usability/ The conclusion: "Always include a visible text label. As Bruce Tognazzini once said, 'a word is worth a thousan…

I think, while icon + label is good for things like UI buttons, this is not that great for tables. There are a few reasons for this:

- icons really work only as a small set (e.g. checkmarks versus crosses or similar) – so the only useful application is also in a context which is already highly selective and with low ambiguity. (Familiarity helps and should be a major concern.)

- icons are useful for scanning for rows with a given property – however, to scan icon + label we must not only scan horizontally as well as vertically, but have also to scan items with varying column alignment, which isn't favorable for a vertical scan, either.

So, if it is a case for icons, say, each row has a state out of a set of three, there shouldn't be need for an accompanying text label (this can be learned from a tooltip or a legend), which is also apt to destroys most of the advantages of having icons in the first place. Having both just introduces a mode change, which isn't helpful for scanning, either. If you must have both, put them in separate (adjacent) columns. (In this context, we may note that tables are really about reducing redundancy, though.)

Moreover, to be meaningful in tables (or menus – yes, I'm speaking of recent macOS), icons should only be used to represent state or to represent actions that modify state. There are, of course, exceptions to this, but this may be still a viable starting point…

Re: Stop using icons in data tables

#55
post #12

The title is misleading. The message is really: Stop using unlabeled icons in data tables. It says, "Norman Nielson argues that text + icon has the highest cognitive recall and lowest error rate" Here's what the Nielsen Norman Group says about Icon Usability: https://www.nngroup.com/articles/icon-usability/ The conclusion: "Always include a visible text label. As Bruce Tognazzini once said, 'a word is worth a thousan…

I was confused what OP is talking about, becuase in all the screenshots I don't see any unlabelled icons, I just see icons next to words. I was wondering... is OP in fact saying it should be just words, without the icons next to them? Which is an opinion you could have, I think it's not entirely unreasonable (although as you say Nielson contradicts), but I wasn't entirely confident it is the point OP is making... or…

OP is talking about the third example image, where there are these "stars" icons in the title row of the table. It does look like what people are using to mean AI-something, but indeed.. AI what?

Re: Stop using icons in data tables

#56
The same could be said for color, just to then go into detail that using color codes on too many columns and with unclear associations are the problem. This is just a weird way to frame an article. Colors and icons are indeed essential to quickly parse and use big amounts of tabular data. All the criticisms the author brings forward are specific problems of some implementations or just poor choices. All this is solved by putting a few key icons in their own column next to the text field they belong to, so users can hide and show them as needed depending on the task. I have eg. a basic indicator that is a red exclamation mark or yellow question mark or green checkmark and another one with a colored icon representation of a category field. Its 100 times easier to get a quick impression and find something than without those. if someone has a problem its 2 clicks to hide the columns, thats the universality of a table ui! Don’t let absolutist click bait articles fool you.

Re: Stop using icons in data tables

#57
post #55

Earlier quoted context omitted.

I was confused what OP is talking about, becuase in all the screenshots I don't see any unlabelled icons, I just see icons next to words. I was wondering... is OP in fact saying it should be just words, without the icons next to them? Which is an opinion you could have, I think it's not entirely unreasonable (although as you say Nielson contradicts), but I wasn't entirely confident it is the point OP is making... or…

OP is talking about the third example image, where there are these "stars" icons in the title row of the table. It does look like what people are using to mean AI-something, but indeed.. AI what?

It doesn't matter if the icon is ever so slightly ambigious compared to other systems, the label text next to it removes any ambiguity and makes the message perfectly clear, as long as it is consistent within a product. To any newcomer to a new system, most icons don't make sense, so the text next to it is an invaluable hidden tutorial. OP didn't stop to see that the circled checkmarks look like clocks, so it just highly opinionated.

Re: Stop using icons in data tables

#58
post #8
post #6

I disagree completely. I have hard times parsing the text. Simple icons are a life saver for me in big data tables

What if you don't know the icons yet? And wouldn't colors be just as helpful? Anyway, they should design for the majority. Or use some configurable approach, like with CSS, to make everybody happy. Opinionated is so last year.

What if I didn't know the words yet? Usage of totally unintuitive icons is obviously wrong. Colors can be helpful too. Colored text - not as concise as recognizable icons.

Another non-negligible advantage of icon is, that it is language agnostic. Not everyone is fluent in language they have to use.

Re: Stop using icons in data tables

#59
post #55

Earlier quoted context omitted.

I was confused what OP is talking about, becuase in all the screenshots I don't see any unlabelled icons, I just see icons next to words. I was wondering... is OP in fact saying it should be just words, without the icons next to them? Which is an opinion you could have, I think it's not entirely unreasonable (although as you say Nielson contradicts), but I wasn't entirely confident it is the point OP is making... or…

OP is talking about the third example image, where there are these "stars" icons in the title row of the table. It does look like what people are using to mean AI-something, but indeed.. AI what?

What is the first image supposed to demonstrate? Very confusing post to me.

Other people in the thread I see are arguing the OP is saying the icons next to text would be better off as just text without icons, the icons still make things messy and increase cognitive load. (which again, I think is worth considering at least, sure!) But you don't think it is?

I think I'm just going to put this away concluding it's a very unsuccesful article at whatever it's trying to do.

Re: Stop using icons in data tables

#60

I literally can't tell what the author is arguing against or for. All the example table images seem fine, and have no captions saying whether they're supposed to be examples of good usage or bad usage. So either I have no idea what "bad" examples of icon usage are because the author doesn't show any, or the author thinks some or all of them are bad when, to me, the icon+text+color examples seem great (and one figure…

Hey, author here, I'll attempt to clarify.

In a data grid or table the relative cognitive load of the page is already very high. Adding iconography to the table body content is often unnecessary and increases visual noise, processing requirements, and generally reduces readability/scanability.

I've always felt that icons in this context are a risk or liability instead of a strength. I decided to info dump my findings to my team then published it as an article.

I probably could use a good editor to help me next time!

Post reply on HN