[Salathe]

Nowadays, we're all used to hearing that a basic requirement of coding up a page in HTML is that it must validate without any warnings or errors. But that often seems to be the only real requirement in so far as HTML quality is concerned.

When writing your own HTML, do you code using tables for layout; do you use lots of divs with some headings and paragraphs thrown in; do you add in lots of hooks (ids, classes, etc) for CSS to grab on to; do you spend 20% of the time writing the bulk of the HTML and 80% tweaking it to a work of art?

[Ubbe]

Well after getting used to and comfortable with xhtml and css, I think it's far more effecient than any other method, and thus also less time-consuming.

Tables - are just a mess lol.
invalid html - will usually give you a lot of cross-browser problems so you end up spending more time with it anyhow.

[Jamie]

I think if a HTML document is valid it should be cross-browser supported but unfortunatly that isn't the case. Why can't they do something about this? Sometimes it can be a nightmare.

But i put in a fair bit of effort into coding, if it's for a client i'd usually work on it a lot harder.

Jamie

[Sam Granger]

I love tableless and almost always code without them. Unless it's for tabular data, then I always use it for that. But the cleaner, the better IMO. Only thing I sometimes struggle with is li menu's that stretch with text (rounded corners).

I always code validated code for clients.

[Morgan]

Nowadays, we're all used to hearing that a basic requirement of coding up a page in HTML is that it must validate without any warnings or errors. But that often seems to be the only real requirement in so far as HTML quality is concerned.

In all honesty, I would have to disagree with that statement. Markup validation, which can easily be overlooked, is not the only requirement for HTML quality, as to all developers, their own style is correct. Quality markup can comprise of various aspects, such as the cleanliness of structure and the load base. I will give you the benefit of the doubt however, that validation does play a large role in such a case.

Tables - are just a mess lol.
invalid html - will usually give you a lot of cross-browser problems so you end up spending more time with it anyhow.

I also disagree with your statements. Tables, which are still compliant, may not necessarily be a mess if the content "needs" it. As long as it remains clean and functional, it still has its uses. However, table-less markup does allow for more fluidity. In terms of the cross-browser problems, even validated markup can. Browsers all parse differently, so what may seem compliant in terms of validation, it can just as easily be viewed differently.

As to the initial question of how much effort put into HTML, I can honestly say quite a lot. As I am a beginner to web specific development, I tend to include excessive content, which, in the long run, makes it more difficult to effectively transition into a more stable situation with xHTML and CSS.

[Diffraction]

I never use tables for layout. All my code is Valid XHTML 1.1.

[the matt]

I Always code tableless & valid. After I code the layout I go through and check to make sure its compatible with other browsers.

[Will Roberts]

Valid coding is a must for me but that doesnt mean it has to be valid.

Go take a look at Google and try to validate that
Clicky