Re: [MACCAWS] Software Test Regimes

--- Chris Kaminski <[email protected]> wrote:
> On Monday, July 22, 2002, at 11:33 AM, Rick [Kitty5] wrote:
> 
> > I have to agree with this, hacks and cheats may be bad but 
> > sometimes they
> > provide the only soloution. My clients dont care how I arrive at
> the
> > finished product, only that I do arrive and that it meets spec.
> 
> If they even care about the spec....
> 
> > While there is a need for hacks I suggest we support them,
> perhaps even
> > going as far to explain any pros and cons.
> 
> I'm not sure there's any reason to even mention 'hacks.' I mean, 
> either a document validates or it doesn't, no? If it validates, I'd
> 
> say it's a noop as far as this particular group is concerned--at 
> least at first.
> 
> Maybe the FAQ gets some sort of "whenever possible, developers are 
> encouraged to adhere to the spirit as well as the letter of the 
> standard" and have done.
> 
> In v. 2.0, specific hacks can perhaps be evaluated based on how 
> 'damaging' they are to the goals of standards adherence (i.e., how 
> badly tehy violate the separation of logic/structure/content). To 
> start, though, that sounds like a great way to start some religious
> 
> wars, and even assuming everyone here is professional enough to 
> avoid them (I know I /never/ get drawn into such sillines, ever; 
> especially not about free stats on the evolt or css-discuss 
> lists ;-), it would still be rather time-consuming. I think that 
> time could be better spent on research to make the business case, 
> rather than on focusing too much on the technical details.
> 
> 
> chris.kaminski { functionNewMedia :: <http://www.function.de> }
> 

I'm wondering if one thing we can focus on (I think I mentioned this
already) is the separation of content and presentation. For example -
ok, you've used a table to set a 2 column layout because for what you
are trying to do, you simply cannot get it to "work" in CSS (not a
great example, just bear with me). In that case, we might say - fine
- but the table must NOT include any presentation code. It's merely a
container. Then at some point, when whatever browsers catch up with
whatever they were trying to do, they can simply replace the table
with CSS. If its done right, its very little work, and the new code
slides right into the rest.

I believe this would even validate? On a side note, I don't think I
ever noticed until yesterday that zeldman.com uses tables. I found
that interesting.

Tom


=====
var me = tom.pixelmech.webDeveloper();

http://www.pixelmech.com/
http://www.maccaws.com/
[Making A Commercial Case for Adopting Web Standards]

__________________________________________________
Do You Yahoo!?
Yahoo! Health - Feel better, live better
http://health.yahoo.com