CSS Layout issues in IE7 - overlapping borders

[leveretdreaming]

@donboe

noonnope was just ahead of me.

But I guess that you’re mixing things up. In one of your posts you talk about separation of content and presentation? Which has nothing to do with a semantic vs structural approach

Ok, I think you’re right - I’m tripping over terminology.

What I said was something like
‘…your implementation is changing the content to suit the presentation…’
which isn’t quite right.

Instead what I was really trying to get at is something like:
‘…your suggested implementation throws away the semantics of the markup in order to minimise the number of tags used…’

[system]

@leveretdreaming

Um… the exact opposite of what you said.

The purpose of div and span is to allow content to be grouped together into meaningful chunks. The fact that they do not have meaning in themselves means that they can be assigned any meaning which there isn’t already a tag for, by giving them a meaningful class or ID.

Really??? But I see EXACTLY what you’ve said:

@leveretdreaming

I disagree: these divs have a semantic purpose of grouping together the heading and paragraph tags into sections. These groupings genuinely are independent sections and without those div tags they are just collections of text […]

[…]

Ok, this I do have a problem with: Those would not be semantic divs (teh acid test is what id’s would you give them, other than “leftcol” and “rightcol”?). The implementation you suggest does not group content together into meaningful chunks, which apart from anything else makes the code less readable.

^ Ahem… beginner showing off… ahem… The content needs not to be grouped, but structured. Big difference. And divs are not the only ones doing the structuring. And divs are NOT semantic. And semantic attribute values is not the same with semantic html element use. And so on and so forth

---

@leveretdreaming

(I’ll quietly ignore the “beginner” remark)

Please don’t, I’ll stand behind it proudly And you’ve just committed a paradox. You didn’t quietly ignore it

---

@leveretdreaming

Ok, I can’t see us getting any further on this particular point: I Know I’m right on this point, and you presumably Know you’re right as well.

Are you the same guy saying:

@leveretdreaming

Besides which that was an unnecessary cheap shot.

---

@leveretdreaming

This is no longer required. It was used to give an ajax call something to populate. If I had needed to keep using it, I suppose the solution would be to insert something into the DOM on the fly (probably easier in Jquery))

I fail to see how this will make it semantic to begin with. A p having a class flash is a semantic nonsense in my book. But hey, only presumably

[leveretdreaming]

Dave, all good points and food for thought. Just a couple of comments.

@DaveMaxwell

Yes I do, and if you were going to HTML5 validation, perhaps I’d buy into it but you’re not.

I disagree. My aim is to mark up the distinction between:
(header1 text1 header2 text2)
…where header 2 doesn’t make sense outside of the context of following header1

and (header text)(header text)
where the two sections are independent.

@DaveMaxwell

Kind of like the directional markers on your front page. I had no idea what those meant until I hovered over them with my mouse - that’s an important piece of information that I shouldn’t have to do extra work to get to - what if I visited your site with my smartphone which doesn’t do mouseovers? Oops…a small key somewhere on that page would have been nice.

Yes, totally aware of that (separate) issue and will be addressing it in a future release by replacing the compass point with part of the postal code (which everyone in London knows and recognises.

[leveretdreaming]

nonnope I don’t think you’re prepared to engage properly: you seem to be mainly trying to pick holes in my use of language.

It doesn’t seem worth the effort required to respond.

[leveretdreaming]

I’ll leave you with the following article:
http://www.thehandcoder.com/markup/the-difference-semantic-markup-vs-structural-markup/
Not sure exactly who it’s agreeing with, but it’s interesting in any case.

[system]

Right, right, I’m the bad guy

Just a reminder:

@leveretdreaming

Ok, that’s fine, it’s difficult to understand, but if you think about it it’s necessary to make the layout work correctly:

[…]

In any case, this is not an example of your assertion that I am not correctly separating content from presentation: it’s just difficult CSS.

[…]

??? This is standard semantic markup: the purpose of those divs is to group content together into independent sections.

[…]

This is not separation of content from presentation, but some sort of misguided “optimisation” of the markup to suit a particular layout.

Please correct me if I’m wrong, but you seem to be operating under the serious misconception that best practice is to only use tags which are absolutely necessary to achieve the desired layout.

Thanks for references. Good luck with your web page. Honestly! Here’s mine, straight from the horse’s mouth:

[URL=“http://www.w3.org/TR/html401/struct/global.html#h-7.5.4”]<!ELEMENT DIV - - (%flow; )* – generic language/style container –>

I can’t see no sections, semantics or anything like that in there. How did you came to those conclusions about div, beats me

div is “the generic container of the last resort”. First, you need to know enough HTML as to see if there aren’t any other more fitted and semantic container elements for your content. Slapping divs carelessly around means insufficient HTML knowledge. Beginner posing as a pro

PS: div is semantic-less. divitis: cure it.

[leveretdreaming]

Thanks

[xhtmlcoder]

I’ve been called-in; since obviously there is some misunderstanding about markup. Yes, people can go overboard with application of DIV however they are designed to hook ‘styles’ or ‘languages’ are they are generic containers.

Presentational markup is considered bad practice; as is placing unnecessary title attributes repeating the exact same thing as the hyperlink itself.

Having an EM in a H1 is also a little weird since it is just overkill, over emphasising the already most important heading in the first place.

I’d agree with Dave on some of those points about stripping some of the unnecessary DIV wrappers.

The superfluous DIV values do not really add semantic structures to the page they are neutral and like I previously said are; generic block-level language/style containers. Nothing more… if they were more then also SPAN would be completely redundant.

They don’t create section islands; your headings denote the sections like was mentioned. Even if you buy into that non normative HTML5 mumbo-jumbo; they realised you cannot use DIV to define semantic sections.

The DIV is no more relevant than the P paragraph element. The DIV element is a generic container for flow content that by itself does not represent anything.

[leveretdreaming]

nonnope: yes, I would consider someone who seems to take pleasure in antagonising others and putting them down a Bad Guy. Perhaps I’ve read you wrong, but please look over what you’ve written and see that that’s how you come across.

If I have come across like that also, then I apologise, this was not my intention.

That article was not my “reference” - just something I found during the discussion.

[xhtmlcoder]

Off Topic:

Don’t worry, Noonnope’s English is improving; I’d assume his first language is Romanian. Sometimes people get a little confused by what he is saying and take it the ‘wrong way’ he wasn’t trying to insult you.

[system]

Off Topic:

Yes Robert, Romanian