If you're using a low contrast/minimalist design, many users won't be able to see what is important! There is a minimum ink you should use in your data to ink ratio:
4.5:1 contrast ratio for text and 3:1 contrast ratio for geometries (non-text).
Speaking of contrast, make important elements even higher contrast. Create a hierarchy using size, boldness, or color contrast to guide the user through each step in your graphic.
Test and validate the whole graphic and all its little components work in harmony.
We (the creators of charts) might visually gestalt a pattern non-sequentially. We are trained in our own process.
But folks with cognitive disabilities, rushed executives, screen reader and keyboard-only users (etc) need a straight path.
"Provide and guide" w/ sequential access
Your interactive features are not very robust.
People with disabilities need to be able to use that brush filter somehow. Even sighted, mouse-using experts can easily misuse them.
Use standard UI elements (before and alongside fancy ones) to perform complex interactions.
You might be tempted to create something cool out of a combination of other things (mashing different chart types together or solving multiple tasks in a single interactive chart, etc)...
but use clear, standard experiences first before inventing an over-featured, flimsy, mess.
Speaking of interactive features... don't make something that looks like it is interactive when it actually isn't! (This image is actually a waterfall, not stairs.)
Use semantic html when creating anything interactive! And of course: remove interactive features that do nothing
Lastly (and most important): please don't add accessibility too late. It will be so obvious that it wasn't part of your original design. While better than not doing it, you'll bloat your experience and make things look amateurish.
Start doing accessibility now!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
So IDEO largely built its marketing around the idea that The Designer is a higher-level thinker who can sift through all of the mass-produced junk that was rising up as a result of automation and find quality.
This was rooted in racism and was doomed to fail... (thread)
I saw some amazing talks and met some amazing people at #CHI2023.
I want to thank everyone who was masked: it made it easy to find folks who really care about disability.
And I can confidently say I made some really wonderful friends, finally connecting with so many of you!
It was very funny to hear the exact phrase "Frank from Twitter" so many times!
And while I have loved remote work (seriously, my physical health completely changed for the better when the pandemic started), meeting folks in person has actually been wonderful.
I set out with a goal to meet with as many folks as possible and learn what y'all were up to. I had 16 meetings with 21 people! Our impromptu visualization + accessibility lunch had just over 30 folks show up!
I actually went to a *party* lmao!! (I never do this.)
I think that folks like @AnhongGuo and others stepping in to provide more feedback, support, and help is key.
And while I hope that @adildsw not only continues to do work, I also hope that the community can do a better job supporting students new to accessibility work too.
No single person gets research right. It might take one person to write a paper, but it takes a whole community to publish one.
I hope that we all can make sure to support each other pursuing future human studies.
👋🏻 Hello #a11y folks, but especially those who build software and games: I have a new piece out where I talk about "Option-Driven Design" and how it is not only a huge trend in accessibility, but surprisingly underdiscussed.
Okay, I want to be as gentle as possible with a review of this now-award-winning work but there are some serious problems we need to make sure we don't repeat.
Our technical HCI class yesterday *just* talked about @elizejackson's "disability dongle" and problematic access work.
Again, I don't want to destroy or attack these researchers! But I definitely want our field to know that there are some things in this paper and project that should have been caught by an accessibility committee and considered in the review process. We can do better.
First and foremost is the language in the paper (and use of "insane" in the tweet about the paper). It is important to use inclusive language.
Do not refer to users with disabilities as "suffering" from a disability ("suffered from low vision"). This is not okay!
Absolutely amazing folks will be there, including my past collaborators and friends @lnadolskis and @clb5590 as well as legends in this space like Godfrey, Kasdorf, and more.
I'll be on the panel for Session A: "Digging deeper into image and graph descriptions for scientific content" but the whole schedule looks outstanding: