Re: conflation of issues or convergence of interests?

At 15:01 +0200 UTC, on 2007-07-28, Anne van Kesteren wrote:

[...]

> I think it will be easier for features
> to become adopted if they don't require a lot of rethinking, but rather
> can be incrementally deployed. I think that's one of the reasons it's
> important to look how authors are solving problems now.

I fully agree that current authoring practice should be taken into account.
But we should not step in the trap of ignoring the *reason* for a specific
authoring practice. A certain authoring practice may well be due to
limitations of the current spec, or UA bugs. Such things cannot be measured
by merely looking at authors' markup. It may well be that a certain practice
would disappear when HTML5 provides a mechanism that authors find easier to
use and/or is implemented correctly across UAs.

Consider for example the shift from authoring presentational HTML to making
more use of CSS. The HTML spec hasn't change during that transition. UA
support for CSS has.

> I suppose it's a bit like social science. Getting statistics on what's
> currently being done to see how we can improve the situation without
> radically changing it.

I agree such research would be interesting and valuable. But merely looking
at code, without knowing the author's rationale behind it, will give a flawed
picture. It may well be that, for a given authoring practice, authors are in
fact, consciuously or not, waiting for a radical change. (Such as their shift
towards using CSS suggests.)


-- 
Sander Tekelenburg
The Web Repair Initiative: <http://webrepair.org/>

Received on Saturday, 28 July 2007 21:10:26 UTC