Re: flowing around both sides of a float

Brad Kemper wrote:
> 
> You draw an artificial distinction. Commercial authors exist to satisfy 
> consumer needs. Some do so better than others (the better they are at 

Commercial authors are there to influence people to buy products and 
services.  That usually means that they do their best not to create an 
informed consumer base (this is as old as modern advertising, i.e. about 
a century).  Typically the aim is to make people buy their mediocre 
implementation of a product, rather than the competitor's mediocre 
implementation, because they have the prettier (or otherwise appropriate 
emotional connotation - fear may be appropriate for some products) web 
site, even though the prettiness of the web site bears no relationship 
to the quality of the product.

> 
> Yep. And the good design of important sites can aid in usability, if the 
> author/designer took a lot of care to craft it that way, and knows what 
> they are doing. Many people may not want to monkey with those designs 
> too much. Of course you can if you want. I would never seek to take that 
> power away from you.

Unfortunately most sites are badly designed from a usability point of 
view, often through being over designed. People who know how to 
compensate compensate for those sites, with the result that sites that 
are properly designed for the medium may also suffer, and the 
expectations of those who don't compensate are set by the bad sites.

> 
> And there we have it. Thank you for clearing up any lingering doubts 
> some of us might have had regarding your anti-CSS stance. You think 
> people would be better off without it. This really makes me question why 
> you joined the list. Was it just to obstruct the progress of those who 
> seek to advance CSS and make it better at actually styling things?

You misinterpret my position.  Used with a light touch, and the 
understanding that it is only providing hints, CSS is good.  If people 
styled web pages well, as someone with reasonable vision for my age, I 
would have no particular desire to override CSS.

I do believe that the CSS model is wrong if style is more important than 
content, which is often the case in advertising.  In that case, I 
believe that one should accept that one is designing for appearance 
first and use a tool, currently PDF springs to mind, that is optimized 
for reproducing appearance, but also allows the structure to be 
annotated.  I think a lot of the complexity of web design and the 
fragility of web pages is the result of trying to achieve this end using 
CSS.  Bolting more and more onto CSS to make it do a job that is already 
well done by page description languages is not a good way of developing CSS.

Personally, I would be happier with the sort of page that works well 
with HTML and simple CSS rather than an attempt to simulate a printed 
brochure page.

Paradoxically, I find that I end up searching out the PDF pages on web 
sites because they are not heavily designed and they do contain a lot of 
real information.  That seems to reflect a fundamental confusion between 
the roles of HTML and PDF!  We have a situation where the tool that is 
intended to reproduce marketing literature exactly is actually used for 
technical documentation and the tool that is designed for documents, is 
used to produce pretty pages.

CSS provides a way of stopping HTML being abused as a page description 
language, ignoring its proper semantics, but it should not be seen as an 
alternative to a proper page description language where the priorities 
favour such a language, which is the case for a lot of commercial web 
pages.  That doesn't excuse the need to consider non-visual users and to 
allow tools for users with cognitive difficulties to identify controls 
correctly.

> 
> You may wish to consider involvement in the HTML WG, if you are not 
> already, in order to help define the proper use of HTML and making sure 
> it evolves to satisfy your semantic ideals even in the absence of the CSS.

This is more and education than a language design problem.  You have to 
convince authors to think about content rather than form, and you have 
make the public aware that something better is possible; people's 
expectations of the web are dictated by how they see it used.

Whilst it is always important that HTML should work with no styling 
applied, CSS can then make that more pleasant to read, but the CSS must 
be done with a light touch.

My main problem in joining working groups, though, is that I have to do 
all this in my spare time, and I don't even really have time to do this 
list justice.

I'm a content consumer who happens to know a lot about the delivery 
technology and history.
> 
> <sarcasm>Yes, clearly giving designers the choice of using color or 
> specifying fonts was a big mistake.</sarcasm>

It appears to have been :-(.  Part of this is due to inexperienced 
designers, moving into the new technology and simply playing with the 
capabilities.  Note that I do normally have colours enabled, and only 
turn them off for particularly bad sites, but those sites have turned 
out to owned by very big organizations.  However, the fad for font sizes 
that are too small to read easily and too small to be handled by font 
scaling on visual displays (most degrade to the minimum recognizable 
matrix of about 7x5 that went out of fashion in the early 1970s as 
technology improved) has been around for several years now, and doesn't 
seem to be going away.

Incidentally, an original TBL paper said that colour had no place in 
HTML!  It's often forgotten that tools for advertising copy on computers 
are not new, but pre-dated HTML.  HTML originated as a universal means 
of communicating information, which was a niche rather different from 
the current commercial use of it, which was already covered, except for 
the internet aspects, at the time.

I would prefer my documents to be rich in content and consistent in 
style (i.e. not varying too much from the browser styling, so that I 
don't have to keep deducing the design paradigm), but if people are 
going produce documents that are rich in style (and typically low in 
content) I would prefer that they used tools that were unashamedly 
presentational, than use a tool that is about content with layers and 
layers of flakey styling added.
-- 
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.

Received on Friday, 4 January 2008 09:09:29 UTC