[nailer]

> Respectfully screw making users rely on AI for accessibility.

Why? It's the right tool for the job.

> Just make the damn page accessible already.

Oh so just modify every website and expect the disabled people to wait while this happens?

This disabled web browser industry doesn't care about disabled people. Their solutions don't work, disabled browsers are expensive because government grants are given to purchase them.

[jakelazaroff]

> Why? It's the right tool for the job.

No, it's not. Why should disabled users be forced to indirectly interact with a webpage via a non-deterministic agent, rather than directly interact with one that's specifically designed to accommodate them?

[nailer]

> rather than directly interact with one that's specifically designed to accommodate them?

Because a world where that happens consistently doesn't exist, it hasn't existed for the last 20 years we've been using ARIA tags, and won't ever exist.

[jakelazaroff]

Your advice to "avoid aria tags" would make that a self-fulfilling prophecy. The ways to make it happen:

1. A robust set of web primitives that are accessible by default, and

2. A government that will actually enforce laws (which already exist!) requiring websites to be accessible

[nailer]

> Your advice to "avoid aria tags" would make that a self-fulfilling prophecy.

As mentioned ARIA has had 20 years to succeed before my Hacker News post. ARIA will continue to fail with or without me.

[jraph]

As you mentioned, without proof. Sorry, but I think you have it completely backward.

ARIA attributes are only one of the tools to help with web page accessibility and are somewhat last resort when you can't do what you aim to do with bare HTML.

The first tool is to not break stuff in the first place.

The solution to "accessibility is not ideal across the world" is certainly not "Outright avoid tagging stuff for accessibility anyway", as if using ARIA attributes were somewhat harmful. It's not, unless you misuse them, and no, the spec isn't unworkable, and you also don't have to use it all.

The response to "software is broken" is not "software has had 50+ years to be bug free, let's put the burden on the users to deal with it since obviously developers can't do bug free".

[nailer]

> As you mentioned, without proof.

I feel that claiming the web now consistently complies with ARIA is an incredibly bold claim, and the burden of proof is on you but you can test the top 50 websites yourself if you genuinely believe that.

> are somewhat last resort when you can't do what you aim to do with bare HTML

This thinking was popular 20 years ago when ARIA was created. Application-like behaviour, which nearly always means JS, is the majority of websites.

> ARIA attributes were somewhat harmful

Wasted engineering effort on minimally effective outcomes is harmful.

> "software has had 50+ years to be bug free, let's put the burden on the users to deal with it since obviously developers can't do bug free".

Others not following your religion is not a defect.

[jraph]

For a user running into broken pages, sure, you have to compose with what you have.

As a developer, however, get your shit fixed! And that fixing doesn't involve any MCP. Don't expect visitors to run AI...

[nailer]

> Don't expect visitors to run AI...

Why not? That's a more reasonable expectation than asking every website to change.

[jraph]

Because ethical issues aside, that's expensive to run, unreliable, non-deterministic hot mess that doesn't have predictable behavior. And predictable behavior is somewhat key.

I won't blame a disabled user seeking AI-based tools to browse the web to survive with what they have, but I will totally blame the devs who created the situation.

[nailer]

A user interpreting visual meaning is vastly more reliable than expecting all the websites the user interacts with to be ARIA compliant.