I'm curious if you used a particular tool for this comparison or a LLM. I ask because (layered irony) I'm making a general tool for multi-dimensional item comparisons, using Claude to build the tool (but not the comparisons, not that I mind).
The existence of non-profits, and the possibility that they might be net-beneficial to society, is counter to the dominant religion of the day. Profit motive = good, other motive = bad. Welcome to UChicago.
I'm a big proponent of use of design systems for efficient remediation and maintenance, so I'm also aware of how a small oops bug can have wide repercussions.
Let me know if you'd like to chat about use of your system to help enforce proper use of semantics and ARIA in web UI. It looks good.
From your title my immediate thought was "cool, maybe this will move us a bit closer to making components (or the testing thereof) cover accessibility thoroughly by default".
See, the idea with the semantic web, and the ARIA markup equivalents, is that things should have names, roles, values, and states. Devs frequently mess up role/keyboard interaction agreement (example: role=menu means a list will drop on Enter keypress, and arrow keys will change the active element), and with ensuring that state info (aria-expanded, aria-invalid, etc.) is updated when it should be.
Then I checked the Antithesis website. They don't even have focus state styling on any of the interactive elements. sigh
Without negating your point I want to add that at some threshold of tediousness, usability issues become accessibility issues. The fact that this threshold varies from individual to individual makes heuristic guidelines difficult.
> as opposed to using AI as a total replacement for your thoughts or your writing and therefore removing the personal touch?
I'm really having trouble grasping the true breadth of this problem in the wild. How much of it am I not seeing because the mods filter it out first? How much is faulty signal detection from readers?
Let me refer you to my buddy Anton, a software developer in Ukraine. He has CP and it makes typing and communicating by speech very slow and tedious.
https://www.youtube.com/shorts/aYbDLOK14uM
IMO his writing style is quite melodramatic. I have asked myself, how much of that is his perhaps overly compensatory tendency to project an articulate voice, and how much of it is applied by his AI tools?
The last time I saw Anton in person I asked him about his writing process, and he said something like, "I just draft it and then ask ChatGPT to make it sound professional or whatever." So after thinking about it for a while, I have decided that this is his preferred voice, so I'll accept it as his voice.
IMO it is not for you to decide how people recast their own voice. Once you adopt that dogma, you're committed to denying other people's experience of discrimination (through the lens of disability's symptoms). Whether or not you participate in that other type of biased discrimination is irrelevant.
> Listen: every idea you've ever had, every single one, some cocaine-addled sales critter has had too. And they're better than you at SEO.
On most days I am resistant to stereotypes about "the welfare state" eroding incentives for entrepreneurism and innovation. But if you're going to wave it in my face like that, I might have to reconsider.
I bought a Dell Precision 7910 2x Xeon E5-2687W v3 (10 cores, 20 threads each) with 32GB RAM and 512GB SSD for $425 including shipping. I found that Windows 11 Pro will recognize only 20 of the virtual cores/threads. I don't feel a need to upgrade to more expensive Microsoft OSs at this time, so I just run Ubuntu natively on that box, which recognizes all of it. Assuming used DDR4 RAM returns to more reasonable prices at some point, I intend to load that box up to the 768GB max.
I'm always happy to see more innovation in this area. It'd be great if you could make your model, weights, and training corpus public (preferably under a permissive license) on GitHub. It'd also be great if you could run some benchmarks against the other similar tools in this area (I'm thinking particularly of Mathpix, Equatio, and Microsoft's math OCR in OneNote, Word, and Azure APIs. If you make your test corpus and code available I could set up the benchmarks for you.
Just this week I was working on something similar but specifically for users who have disabilities, so they can more easily report issues to site owners. I also combined general annotation capability so other users (of my browser extension) can read their comments. And also compatibility with Hypothesis (https://github.com/hypothesis, https://hypothes.is), also using the W3C Web Annotation spec.
I hadn't thought of the drag-and-drop bug metaphor; I like it.
I had also considered recording mouse and keystroke events up until the time that the bug is marked, and then bundling those events (sanitized) with the bug report for more precise repro steps, but of course that's a bigger ask for the opt-in.
You managed to make a web app in 2025 that has almost no HTML semantics and almost no keyboard accessibility (so very clearly and deeply not WCAG-conformant). Gosh, I haven't seen that in quite a while. Mind sharing which tools you used to make it?