HTML Markup Validation: Your thoughts.

[The Alien]

Interesting discussion, but it's currently very much biased towards people with a disability who need some (software-)tool to understand images. This was first posted in Remi's first post here. The link points to a webpage where someone is holding a speech to prove his/her point of view about why people should write correct HTML code. People with a disability certainly deserve this, but there are more reasons than that. On page 2, I see more and broader reasons why people should write correct HTML.

But if I were to continue on the subject of people who use tools to read the alt tags of images (since everyone here loves to discuss this issue ), my opinion would be this:
We've all heard of the saying "a picture speaks a thousand words". Well, this is true. Now, try to put 1000 words in the alt tag...
As an example, I'll take an image from a recent thread the Xara Gallery.
Here's the challenge: what would you put in the alt tag? It could be:
- The Green Goblin
- The Green Goblin on his hoverboard
- The Green Goblin on his hoverboard laughing diabolically
- The Green Goblin on his hoverboard laughing diabolically and holding a pumpkin bomb
- The Green Goblin on his hoverboard laughing diabolically and holding a pumpkin bomb drawn in a cartoonish fashion
- The Green Goblin on his hoverboard laughing diabolically and holding a smoking pumpkin bomb, drawn in a cartoonish (dithered) fashion
- The Green Goblin on his hoverboard laughing diabolically and holding a pumpkin bomb drawn in a cartoonish (dithered) fashion with a blue/yellow faded background with white stars and the title 'GREEN GOBLIN' at the top of the image in bold letters (in green, of course).
- ...

Chances are, most webdesigners will go for the first or the second (maybe the third) alternative in the list. These choices are definately better than nothing, but they're not anywhere close to adequate IMO. The user won't know what this picture is displaying beyond that it's the green goblin on his hoverboard (if he was born blind, he wouldn't even know how the green goblin looks like). The user wouldn't catch the image's atmosphere and the whole thing would look like a simple composition without any depth.
Now, you can say that the user would have to use his imagination to visualise the character, very much like reading a book without pictures. But this goes for artistic pictures like those posted on the Xara gallery. What about webpages that have an image of some sort of schematic, similar to what you may find on e.g. Wikipedia? How on earth would a webdesigner come up with an alt text that suffices there?
We can all behave like good webdesigners and put alt texts inside our img tags, and sit back comfortably in our chair knowing that we've done our jobs right and that the minority with a disability can also understand ( ) the image. But the fact is, that won't help a lot.
I'm sorry, but alt texts are only sufficient for the most simple images
For more complex ones, one should always use a caption - like figures in scientific papers - to explain what's going on.
Another solution would be in hoping that the technology would advance to a point where images could somehow be "shown" to people with visual disability using other methods (tactile, auditory, maybe even by some direct chip-to-brain interface).
These are the disadvantages of alt texts. We are not robots, we're humans who should use their common sense and good judgements. If you think alt texts won't help, don't use them. Instead think about what would help and what the users need and change your plan accordingly. Choose what users you want to focus on.

Anyway, as I said, there's more to making your code valid than just taking care of people who use other means to communicate with the web :\