Conversation

Notices

1. Embed this notice
   Stefan Bohacek (stefan@stefanbohacek.online)'s status on Wednesday, 08-May-2024 00:33:26 JST Stefan Bohacek

   I honestly enjoy writing good, descriptive alt text. Once you get used to it, it's no longer a chore, rather a fun exercise.

   It gives you an opportunity to focus on the image you're sharing, sometimes revealing details you might miss.

   It's a great writing challenge as well.

   And it's just the right thing to do to help folks who rely on it.

   #accessibility #a11y #AltText #ImageDescriptions

   • BeAware :fediverse: repeated this.
   • Embed this notice
     ᴚ uɐᗡ (dannotdaniel@mastodon.social)'s status on Wednesday, 08-May-2024 00:33:24 JST ᴚ uɐᗡ

     in reply to

     • Coach Pāṇini ®

     @paninid @stefan oh damn

     I hadn't thought about that

   • Embed this notice
     Coach Pāṇini ® (paninid@mastodon.world)'s status on Wednesday, 08-May-2024 00:33:26 JST Coach Pāṇini ®

     in reply to

     @stefan
     There is a lot of overlap between writing alt-text for images and prompt engineering for generating images using “AI”

     BeAware :fediverse: repeated this.
   • Embed this notice
     botvolution (botvolution@mastodon.sdf.org)'s status on Wednesday, 08-May-2024 00:33:34 JST botvolution

     in reply to

     @stefan
     Sometimes when I've had to describe something slightly unusual or meme-y or chaotic then I've used a sort of Jane Austen wry tone, because why should an image description have to be dry or formal ? Why not try to transmit something about one's personal response to the image? Why not be funny?

   • Embed this notice
     BeAware :fediverse: (beaware@social.beaware.live)'s status on Wednesday, 08-May-2024 00:37:56 JST BeAware :fediverse:

     in reply to

     • ᴚ uɐᗡ
     • Coach Pāṇini ®

     @dannotdaniel @paninid @stefan it's why ChatGPT is so good at writing alt text, I feel.

   • Embed this notice
     BeAware :fediverse: (beaware@social.beaware.live)'s status on Wednesday, 08-May-2024 00:48:22 JST BeAware :fediverse:

     in reply to

     • ᴚ uɐᗡ
     • Coach Pāṇini ®

     @stefan @dannotdaniel @paninid either I use it that way, or I don't post alt text. I'm ADHD and slightly autistic, I cannot for the life of me, explain the textures on the blades of grass in a picture. The AI can do that or it won't be done.🤷♂️

   • Embed this notice
     Stefan Bohacek (stefan@stefanbohacek.online)'s status on Wednesday, 08-May-2024 00:48:23 JST Stefan Bohacek

     in reply to

     • ᴚ uɐᗡ
     • Coach Pāṇini ®
     • BeAware :fediverse:

     @BeAware @dannotdaniel @paninid I have pretty mixed feeling about using AI for this purpose.

     But interestingly, some blind people already rely on it. https://mastodon.social/@botwiki/111477849625070656

   • Embed this notice
     Stefan Bohacek (stefan@stefanbohacek.online)'s status on Wednesday, 08-May-2024 00:56:12 JST Stefan Bohacek

     in reply to

     • ᴚ uɐᗡ
     • Coach Pāṇini ®
     • BeAware :fediverse:

     @BeAware @dannotdaniel @paninid Yes, sorry, that wasn't intended as a criticism, just my personal stance.

     Carry on!

   • Embed this notice
     botvolution (botvolution@mastodon.sdf.org)'s status on Wednesday, 08-May-2024 01:16:02 JST botvolution

     in reply to

     • econads

     @econads @stefan agreed

   • Embed this notice
     econads (econads@chaos.social)'s status on Wednesday, 08-May-2024 01:16:04 JST econads

     in reply to

     • botvolution

     @botvolution @stefan
     As long as it's also descriptive. I saw someone with # altTextIsAnArtForm in their bio, and their alt texts were just jokes which relied on you already being able to see the image. I felt bad for anyone who actually needed them.

   • Embed this notice
     Jupiter Rowland (jupiter_rowland@hub.netzgemeinde.eu)'s status on Wednesday, 08-May-2024 16:55:18 JST Jupiter Rowland

     in reply to

     • ᴚ uɐᗡ
     • Coach Pāṇini ®
     • BeAware :fediverse:
     @Stefan Bohacek @BeAware :veriweed: @ᴚ uɐᗡ @Sampath Pāṇini ® I don't have mixed feelings. From my personal experience, and for my own purposes, it just plain sucks. But that's to be expected. It would require extensive pop culture knowledge plus absolutely extreme niche omniscience in my special field. Like, if I've built something in-world, an AI would have to immediately know everything about it.
     
     I've actually had LLaVA describe an image I've described first. Yes, it was an in-world rendering.
     
     In comparison with my own 25,271 characters (no, I'm not kidding, go check the link), LLaVA's 558-character description was vague, it was painfully lacking and incomplete, it didn't explain anything because it had no idea what it was really dealing with, and it was even glaringly wrong in some points. It would not helped anyone understand the image.
     
     I've got my doubts that ChatGPT can do much better than LLaVA. And I've got huge doubts that ChatGPT can produce something more detailed, more informative, more explanatory and more accurate than me from any in-world image I'd post. After all, ChatGPT would only have the picture while I could look around in the virtual world itself and see things that ChatGPT can't see.
     
     #Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #AltTextMeta #CWAltTextMeta #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta #AI #LLaVA #ChatGPT