• I believe that accessibility helps us to innovate and create.
• I believe that accessibility is something that must be provided.
• I believe that accessibility is different than interoperability.
• I believe that accessibility is not a hindrance to progress.

There is much kerfuffle over Bespin, a <canvas> based tool that was put out by Mozilla Labs. (Incidentally, I’m really hoping it is Bespin as an homage to Star Wars’ Bespin and not BeSpin, as in the conjoining of the two words “be” and “spin”)

Many people are crying foul saying that <canvas> is inherently inaccessible at this point, and therefore the accessibility problem must be solved before the launch of the product/project. This fits perfectly well with the notion that technologies/projects/products need to be accessible from the get go. Generally, I support this sentiment.

Reading the emails, blog posts and twitter reactions has me questioning one particular aspect of this accessibility challenge. The question is not whether or not a technology should be made accessible, but when?

Accessibility Investment

Many individuals and organizations take accessibility very seriously and invest a lot of time and effort into making their work accessible. These efforts and investment are not to be taken lightly; they are precious and should be undertaken wisely.

As an example, when faced with issues of limited budget, time and prioritization, we’ll often suggest that organizations ensure that when JavaScript is on, their applications work properly with assistive technology rather than ensuring that their applications work with both JavaScript on and off. JavaScript on/off is an issue of interoperability — if your app doesn’t work with JavaScript off it sucks for people with or without a disability. Interoperability is best practice, but it doesn’t discriminate based on disability.

So, given that significant investment needs to be made in accessibility in terms of time, effort and money, when is the appropriate time to make that investment?

Consider a couple of scenarios:

1. in dealing with an experimental technology such as Bespin, we don’t know if anyone is going to use it, let alone people with disabilities. What if it sucks for everyone? is there any reason to make that suckiness accessible to everyone?
2. in addition to not knowing if anyone will use it, we don’t necessarily know how they will use it. Accessibility is part of user experience. Simply providing an alternative may provide a basic level of technical accessibility but may be unusable by people with disabilities. I would suggest that it is at least possible that until we know how people are going to use something, we have no idea with the most appropriate alternative will be.

Other emerging technologies such as AIR and Silverlight did not address accessibility in their 1.0 release of their product. Should they have? What if the technology was fundamentally unusable? What if, after 1.0, they looked at the product and said “this stinks, we have to start over.” Would it have been worth the investment in accessibility for a product/project/platform that died on the vine?

Accessibility in Mind and Implementation

Is it possible to include accessibility support “too early?” I’m not saying it should be an add-on at the end of the process/project/product development cycle, but I’m very seriously wondering what the optimal time for integrating an actual accessibility implementation is? Is it enough to keep accessibility architecture in mind from the beginning, but not implement right away? Should we get the basics right first, and then build in accessibility support based on that previously envisioned architecture after we know we have a viable product? We continue to say that accessibility should happen throughout rather than just at the end, but would it actually be better if we left it out,  just for a little while, at the beginning?

Is it a better “business decision” to say very early on “we are committed to making this accessible, but we know we’ll fall short of the mark on our first cut; we want to get this right for everyone, and will, but in order to make it accessible, we need to get this out into the real world to see how people will use it, what they want from it, and then build in accessibility appropriately.”

My feeling — at least right now — is that our job is to ensure that accessibility and accessibility architecture is kept in mind from the outset of a project/product/technological exploration, but not necessarily implemented at the outset.

I’m just throwing these thoughts out there for discussion — there is nothing definitive in here, other than the fact that I don’t think there is going to be one correct answer for this. What do you think?

As soon as they [the user] encounter a wacko result, it reduces their confidence in the search. For some users, this makes it unlikely they’ll continue using Search as a reliable method for completing their objective, since it works unpredictably.

That pretty much sums up my first experience with Cuil, a new search player with VC funding, managed and founded by a few former Google employees that seem keen on changing the search game.

Kudos to them for braving the “search space” or whatever it is called now, exclaiming mightily:

Rather than rely on superficial popularity metrics, Cuil searches for and ranks pages based on their content and relevance.

They completely failed for me on relevance, though as I subjected them to the first test almost any blogger will perform: searching for your own name. While most of the results were relevant, what was associated with them threw me for a loop.

There are several wacko results on that page, including an image of a crying baby beside my name as an A List Apart author, indexing a URL that is a textdrive subdomain. Peculiar.

There is the book that I contributed to, with several other fine authors: Web Standards Creativity, referencing an image that shows the cover of Beginning JavaScript with DOM Scripting and Ajax by Christian Heilmann. Odd. (I mean the result, not Christian. <cheek>Or do I?</cheek>)

But this one trumps all. It is off the wacko result scale:

Now, I loves me some Rob Weychert, but why oh why is his personal avatar associated with my company? Don’t get me wrong. Rob is a very good friend. Along with his stellar beard, he’s a an air guitar genius.

But his two-handed middle-finger salute avatar, ummm, doesn’t exactly send the kind of message that I want to be associated with the company description, if you know what I mean?