So on Tuesday I created three identical posts using three different editors to see if there was a marked difference in the accessibility and structure of the code posterous created from them.
I spent some time on the train yesterday looking through the source, and was quite surpised.
The posts themselves were almost identical, and startlingly the actual content was just one line in 1600.
Just 1200 characters in the 44000 making up the code for one post were actual content. This may not be such a shock to younger readers, but I was schooled in html at a time when it was usual to hand code, and when Dreamweaver was considered by many to be a bit of a cheat. The ratio of content to “fluff” seems at odds with Tim Berners Lee’s idea for the web to be a source of data, upon which we impose the way we wish to see it. Of course I’m an old duffer, and no little of the underpinnings of the modern web, looks like I’m about to start learning.
From my reading of the code I’m inferring that the posterous engine is efficiently parsing the incoming post, and then dropping it into a big bloaty presentation layer. In the review of the site, @ahiggi points out that many sites with interfaces customised by the creator don’t provide a firm structure to gauge accessibility upon, as small changes to the theme can have a big impact on the end user. It feels like playing jenga, if I change one aspect, I’m unsure of the effect on the stack.
SO, what can BS8878 do to help? Well starting tomorrow, I’ll work through the advice and see if it can effectively help me solve my problem. I have a feeling this is going to take MORE than a week, and it will mean learning more about .css than I currently do.
What an adventure.
[screenshot of code block from source for posterous site]