Comments on: An Argument Against Faux Absolute Positioning

By: ian

ian — Wed, 29 Jun 2011 05:15:09 +0000

Now its 3 years later and if your like me your doing mobile ipad and screens which require adaptive /responsive css and flexible containers. Tables in my shop are for tables period. Not layout.

By: Nick

Nick — Fri, 28 Aug 2009 18:30:53 +0000

The only thing you forgot to mention in your article (and I didn’t read all the response posts) is the FAP’s ability to have liquid columns. Tables cannot.

By: Rob Glazebrook

Rob Glazebrook — Sat, 13 Dec 2008 16:37:47 +0000

Debi: I’ve heard a rumor that the latest version of Dreamweaver (CS4) uses a far superior rendering engine to its predecessors. I don’t have a copy myself, so I can’t check (I’m two versions behind now), but if so that might solve your problem.

Can anyone verify that Dreamweaver CS4 can accurately display complex layouts?

By: Debi

Debi — Fri, 12 Dec 2008 19:47:00 +0000

I have read all of the comments on the ALA article & about half of the one here, but to date Eric’s FAP is the only “tableless fluid/dynamic multi column layout I’ve found that works w/o floats – something I need to reserve for content and image layout. Whenever I’ve used floats for multi column layouts, I’ve encountered numerous problems with placing images in newsletter articles, bios, etc. My only problem with it is what was mentioned in the ALA article’s discussion that editors such as Dreamweaver do not show any of the actual content (due to the negative positioning?) until you preview in the browser. I’m redesigning a site & am not sure I want to do the entire site in code view only. Any suggestions?

By: Cristian Deschamps

Cristian Deschamps — Thu, 27 Nov 2008 17:33:38 +0000

I really like your point of vue.

I do semantic and ligher code possible, and my mantra is KISS. Keep it stupid simple.

Like in physics, simple idea always win.

thanks
-Cristian

By: tZ

tZ — Fri, 29 Aug 2008 15:44:22 +0000

I agree with many of things your saying here. However, there are also ways which this technique can be implemented in a purists form. The wrapper is [optional]. You can avoid the wrapper by adding the padding to the items instead. However, if you do this you will need to account for it in your margins. It all comes down to the way the technique is implemented. I’m sure every purists here has build a similar structure just not with the same intension. That is really all that technique amounts to is a structure. If you have a similar structure using block level elements you can easily implement the technique without “unsemantic” mark-up.

By: Suzy

Suzy — Fri, 11 Jul 2008 10:06:51 +0000

It’s not quite new – similar to my own “Flexi-Floats” and http://blog.html.it/layoutgala/

(note: the hacks in mine are no longer required either – I wrote that back in 2003/2004 when NN6 and IE/Mac were still a support issue ;))

Your perspective, might also be a little skewed. When you say that this uses more divs than a table structure, it doesn’t. You include the canvas div as equivalent of the

element.. I would venture that it’s not it’s simply a wrapper div. If you were to have more than one table on the page to gain the non-nested but differing columns/grid look it would be different:

[table class="line"]
[tr class="item" id="item1"]
[td class="sap-content"]content here

and where is a table’s flexibility when it comes to re-ordering the columns (per Layout Gala) or the dragging flexibility the author says he needed

Floats weren’t built for layout anymore than tables were, it’s a hat tip to authors like Eric that at least there’s a choice :)

By: Rob

Rob — Thu, 10 Jul 2008 05:41:36 +0000

Since you did not provide detailed working examples, I can not evaluate your clams. I’d really like to see the code you would use to create the two examples in the ALA Faux Positioning article.

I do know that I can not afford the time to test my designs on every browser under the sun. Since I’m trying to write code that can dynamically generate CSS and XHTML, I’d be quite interested in your technique if it avoids all the browser dependencies and has sufficient flexibility for what I’m doing.

Peace,

Rob:-]

By: Jeffrey

Jeffrey — Wed, 02 Jul 2008 18:22:06 +0000

I work at a company that offers a pretty amazing cms. When im cutting that cms there are some cases that i need to anticipate certain things that a client may do. They are putting in the content not me. I have an idea and a comp and thats it. This technique has actually came in really handy. On clients pages i dont use it on every one i used floats but when it comes to certain things sometimes its tables or nothing and im not going to put a table inside the cms where the client will be putting content.

I understand what you are saying but the few times i have used this i felt a little better knowing i used css and not a table. I have nothing against tables i’ve just went through the everything in a table stage and its not pretty. Even with this extra bit of code no one who i’ve worked with has had any trouble going through the document.

The one layout i did use it for all my content was a strange idea i had when i saw eric meyer slantastic. I used that layout with Eric Sol’s faux absolute positioning. I could then style the elements how i needed them and they would be easy to use in a cms without any overlapping at the footer.

By: Rob

Rob — Wed, 02 Jul 2008 12:51:46 +0000

Great comments so far, all!

I will admit that this sort of a technique would probably be a godsend if you were building a dynamic layout, wherein the pieces had to be movable by the user. I can see where this technique would make things vastly easier there. And I think Chris made a good point where ~removing~ elements from the page would be easier using this technique than with tables.

However, I don’t agree that this technique is better simply because it doesn’t require as much CSS, or as many CSS-based fixes for IE. CSS files are cached: they’re loaded once by the browser, and then not downloaded again until a new version of the file comes along. That means you download a CSS file once, and you’re good for the rest of the site.

By comparison, you’re stuck downloading every page of a website, regardless of how similar in XHTML they are. As such, if I’m going to have bloated code somewhere, I’d vastly prefer it in the CSS than in the XHTML, if purely for bandwidth reasons.