@alyssapanetta
Thank you so much Alyssa!! Everyone was very enamored with your talk, hence the wish to really make sure they got the understanding of this concept right as well.
I’ll pass everything on to the “audience”. They’re watching a full day of talks tomorrow as well.
@kazarnowicz
Ah yes, Mastodon interprets Akkoma a bit strangely and makes tags into links (if they become before other links). But thanks for the heads up. I'l try and remember to not add tags without other links in the page 😅
@kazarnowicz When I myself click that tag I get a tag search on my own instance and all posts with the tag "axecon" from others as well. Don't know why Mastodon does this with posts from Akkoma or Pleroma, but it's a known bug.
Watching Axe-con all day with lots of wonderful people in the accessibility space. Best work day in a long time. Feels like I've been away from reality forever.
https://www.deque.com/axe-con/
#accessibility #a11y #axecon
Insight: HTML email still doing a bad job of accessibility. Zooming in HTML emails works poorly at best. Work on you newsletters everyone, and test them in several clients.
#axecon
Insight: When people are subjected to an overwhelming cognitive load they often have to take long breaks before being able to continue with the task they are doing on your website. These effects are very real and strenuous for a lot of people.
#axecon
Alyssa Panetta's 3 big tips for reducing cognitive load.
1. Semantic HTML and Familiar Layout Patterns
- Maximize germane, minimize extraneous
- Use clear headings
2. Use modern CSS and make your site flexible
- Amazing new flexibility with minimal code
3. Say it twice
- Increases chances of users understanding what you mean
- (my comment: known in communication science as increasing redundancy)
- add motion to grab attention (CSS transforms/transitions)
Article tip:
Modern CSS Upgrades To Improve Accessibility, by Stephanie Eckles:
https://moderncss.dev/modern-css-upgrades-to-improve-accessibility/
And remember: Accessibility overlays are bad!
Read more about Alyssa's talk:
https://www.deque.com/axe-con/speakers/alyssa-panetta/
@alyssapanetta on the Fediverse.
#axecon
@alyssapanetta
Top insight from @alyssapanetta, which I always assumed but hadn't seen hard data for:
The word "Menu" as a button outperforms having a button with a hamburger menu alongside the word "Menu".
The hamburger menu detracts from, rather than improves, usability.
Thank you Alyssa. Great talk! (we are a gang of 40 people watching selected #axecon talks together)
@alyssapanetta
A lot of people seemed to get a bit stuck on the concept of "Maximize germane, minimize extraneous" that you had in your talk.
The words/concepts were unfamiliar for a Swedish audience and we struggled to find an easy translation. Have you expanded on this in text somewhere?
#axecon
“ Australian employer, business tech and research groups have formed an alliance to tackle responsible AI by developing best practice guidance, tools and training for Australian companies using the high-risk technology.
It will offer Australian businesses support with guidance and training on the fundamental challenges of AI like data bias and unknown or unforeseen outcomes so business users build explainability, fairness and accountability into Australian AI systems.”
https://www.inkl.com/a/wawlRjIaBbV
#AIEthics #ResponsibleAI
@Beantin
Yes, instead of making stuff to inflate our own sense of self-worth.
It's confusing to me that designers are excited about building graphical user interfaces (GUIs) using ChatGPT and manage to look past the purpose of the interface they themselves are building.
With the power of chatbots why would people need an interface for buying tickets if they can tell a chatbot to buy tickets for them?
The GUI is in many cases an obstacle to goal completion. It's rarely a goal in itself.
If you see a chat tool as a better tool than the interface of your design software, why would you assume that the same reasoning does not apply to the interfaces you yourself build? They should also become obsolete.
Welcome to the future. 😛
I'm being interviewed about AI and Human rights later today, a follow-up on my talk at Raoul Wallenberg Institute a little over a year ago.
It will be in Swedish, and I think I just thought of the perfect explanatory model. In Sweden we ALL watch the same cartoons every year on Christmas eve (don't ask). One of them is Mickey's Trailer, where Mickey, Donald and Goofy are on a camping trip in a high-tech trailer.
In the revelation scene they all convene at the breakfast table in the still moving trailer, and as they are devouring their food Mickey suddenly asks "Hey, who's driving?!" To which Goofy responds "Why, I'm driving!"
It takes him a while to realise he isn't and in a fit of panic he rushes back to the out-of-control car, inadvertently unhitching the trailer, which then comes crashing down the mountain road with Mickey and Donald still inside, Goofy of course completely oblivious to what's going on.
There you have it. That's the dilemma of AI today, summed up in this question: "Who's driving?"
Oh yeah, if you're curious: you can also watch, or read the full transcript of, my AI and Humans Rights talk from January 2022, which was in English.
https://axbom.com/aihr/
@mensrea
Good example. I know the entire African continent is host to at least 2,000 languages.
And there's the input limitations of keyboards. Thinking of the Ndélélé alphabet for example.
@CatherineFlick @yiningkarlli
Wow, that's one of the scariest reports I've read thus far on this whole debacle 😨 The fact that it was published a year ago (almost to the day) and the researchers were talking about GPT-3 as a potential starting point makes it even more frightening.
Is me!
“A lot of introverts I know, myself included, love spending time in coffee shops working on our laptops or reading a book, because it’s a way to be around people and get that sense of feeling connected to the community.”
https://www.nytimes.com/2023/03/13/well/family/introverts-friendship.html
@akenyg Tack för återkoppling och delning Åke!
Jag suger åt mig av "axbomskt". Jag är jätteglad över att få uppfattas på det sättet.
Alla läser inte. Jag vet. Så för att göra mitt innehåll mer tillgängligt erbjuder jag många av mina texter nu också som ljud. Nej, inte automatiskt. Det är jag som läser in.
Vad bidde det då? Det blev en podcast.
Att läsa in mina texter gör mina texter bättre. När jag märker att något inte känns bra i läsningen så kan jag gå tillbaka och ändra texten. Jag hittar lättare stavfel och repetitioner. Jag får också en helt annan relation till mina ord när jag lär känna dem med rösten.
Och i vissa inspelningar kan jag inte låta bli att leka loss med passande ljudeffekter. Som i det här senaste avsnittet (avsnitt 5) där jag adresserar AI-hajpen. Det du hör här i inlägget är några spridda klipp från programmet.
Det är lätt att glömma att man redan har bra innehåll som man kan förstärka genom att göra det mer tillgängligt. Vilket innehåll har du glömt som du kan hjälpa fler att ta del av? Tänk att du kan nå fler målgrupper med det du har, i stället för att alltid skapa mer för de målgrupper som redan får så mycket.
Om du vill prenumerera på podden hittar du länkar till de flesta lyssnings-tjänster på den här länken:
https://omtanke.axbom.se/subscribe
Eller sök bara på “Omtanke med Per Axbom” i din favoritapp.
@emilymbender @zoeschiffer @caseynewton
Montgomery: “There's a lot of stuff being ground up into the sausage.”
Oh, he came so close to saying SALAMI.
@Kolombiken
hurra vad roligt!
@kazarnowicz @maria
+1 på Account Factory. Jag skulle inte klara mig utan dem.
@simonwilliamson Wow, that’s a great example. I understand that no countries in Africa observe DST. I wish the EU would follow suit.
Daylight Saving Time starts today in the US. In Europe we still have two weeks until we change the clocks. So it’s time for that fortnight when there’s a significant risk for transatlantic scheduling errors.
BUT… remember that it’s not all of the United States that switches to DST. Exceptions to remember are that Hawaii and Arizona do not observe Daylight Saving Time.
BUT… that is not entirely true either. In northeast Arizona, in the Navajo Nation, they do. So, not observing the DST is not true of the entire state of Arizona.
The Navajo Nation territory goes beyond Arizona into states that follow DST, southeastern Utah and northwestern New Mexico. Observing DST allows everyone on the vast reservation to live by the same clock schedule.
Arizona previously observed Daylight Saving Time beginning in 1918, but made the permanent change to Standard time in 1968. Yes, with that exception.
”Too many people are starting to count on systems that inherently can’t keep track of the truth.”
”CNET’s editors learned this the hard way last week, when they published a bunch of machine-generated articles that turned out to be filled with subtle but important errors, with errors in 41 of the 77 articles the bot wrote, all missed by editors who weren’t paying enough attention.”
https://garymarcus.substack.com/p/the-cnet-fake-news-fiasco-autopilot?ref=refind
So this is a cool Fediverse development.
Automattic have teamed up with the developer of the ActivityPub Wordpress plugin… with the intent of Wordpress becoming a ”first party fediverse member” – which I take to mean that integration will be more seamless in the future.
via https://religious.social/@kraft/110003134398099446
@LeoR1010 Underbar. Älskar henne. Tack för heads up.