w3c

[Maljonic]

Is it worth configuring all your page to comply with the w3c validation rules, or is it just a pain in the bum and hardly worth bothering with?

I'd never even heard of this stuff a month ago, but I've been obsessed with getting pages validated - true, it has helped me find some errors that I'm glad to have cleared up; in some cases though, like on a page full of stuff for sale that has a lot of '&' in the code, it just doesn't like it, even returning errors when I change all the &s to &

Anyway, I decided to keep as many pages compliant as possible but not to bother on some pages that are too complicated to validate; just wondered if anyone else bothers with this, and what you think of it?

 

[TdC]

depends what you put in your header. browsers can apply their rendering of the html code to match your header or a "catch-all". I'd do it (though alphanor.org prolly doesn't validate heh )

consider the following: if your code doesn't validate, then there is something wrong with it in the eyes of the people who defined the standard. html can loosely be grouped into a strict adhitherance to the standard and a loose one. you can choose which one to validate in the w3c checker, but you should hold to one or the other, and define it in your head statement.

also, I'm willing to give good odds that at least 80% of all the websites don't give a fig about w3c standards. either by choice, as at some point a judgement has been made that holding to the standard will break some really cool bells and whistles that they want to impress their customers with, or because they simply have no idea that there is a standard.

 

[JingleBells]

IMO, Yes. It makes sure that the pages will load in most browsers and display roughly the same.

because they simply have no idea that there is a standard

I think it is probably that, as a lot of the errors are due to the designer mashing something up in word or some other dire editor.[/quote]

On a side note:

Hutchison-Whampoa's copyright statement:

No person, whether an individual or a body corporate, shall create or establish a hyperlink to the HWL Corporate Website by hypertext reference or imaging without the written permission of Hutchison.

 

[Jonty]

Hi maljonic

W3C standards are just that: standards, which if complied with will hopefully render your pages accessible to the greatest amount of people on the wideest number of platforms. If at all possible, your pages should validate.

However, there's more to life than validation. Having simple, logical, semantic code which fails to validate because of reason x is better than having messy, complicated code which validates fine. For example, a few years back I wrote a site which used tables for layout purposes, little CSS and all kinds of other no-no's. Did this site validate as XHTML 1.1? Yes. Did it achieve the broader goal of simple, accessible markup? No.

I guess the moral of the story is: try to validate your pages, but don't lose sight of the principles behind such standards. That said, not bothering to correct errors in your markup because you have a big document isn't a good reason for not trying

Kind Regards

Jonty

P.S. Standards compliance is still very under used, approximately 99.9% of all webpages do not validate, so think of yourself as joining an exclusive club by making the effort

 

[Jonty]

also, I'm willing to give good odds that at least 80% of all the websites don't give a fig about w3c standards.

99.9% of all sites, and such figures come from standards supporters ... Still, Alpahnor is on the right track

Kind Regards

 

[Maljonic]

Well I am validating nearly all my pages, there's over 200 of them; just a few that are almost impossible to validate and get working properly at the same time, probably only about 10 to 15 at the most, and the message board.

 

[Shovel]

Taking pages which you've written already with no awareness of web standards, accessibility and so forth and trying to make them validate is an uphill battle. You can sometimes make them pass the test, but the actual big benefits that comes from standards compliance (more reliable cross-browser operation, operation on different devices altogether, separation on content, presentation and script) can only really be achieved from the ground up.

So don't worry too much about your current pages, learn from the messages you get back from the validator, maybe consider playing with HTML Tidy (an application which can validate and correct your HTML automatically) and view what changes it produces.

When you start your next site though, keep standards in mind from the outset. Plan your content structure separately from your design and see what happens. It'll be hard, because it's a different way of thinking, but the results are massively beneficial (but I have to go home, so I'll have to come back later and tell you what they are.. ahem).

Ben

 

[Maljonic]

Every new page/site I do now is validated, I just go through the old pages to pass the time when I'm a bit bored; I quite like it, bit wierd like that.

 

[Shovel]

Well, this is how I plan/make a site these days (and the reasons why).

1= Sketch designs on paper (a lot of people use PhotoShop or Fireworks for this, I tend to move to them later)
1= Take some dummy content and arrange the linear, logical HTML structure. Start with the <h1> for the entire site, search and maybe some major navigation (three items at most), plus a link to the full page nav. Then an <h2> for the content heading, with the content following. Then the navigation, then the footer.
You end up with a nicely structured document. No 'layout' of sorts, no presentation apart from the browser default. You also don't have any <div> or <span> tags to hook into. This, ultimatley, is what it will look like on a simple platform like a mobile phone.

2. Based on the sketch/image I've produce I can identify which parts of the page need grouping in a div. Thus a navigation bar gets a single div around it, if necessary the content will too. Footers normally get wrapped too to aid in the whole "get 'em to the bottom" malarky. This varies, of course depending on the design.

Currently, CSS isn't quite powerful enough to completely avoid using non-semantic elements like <div> and <span> to hook into for layout. CSS3 has some fix for this, but that wont get browser support for a good number of years.

Working in this order though, I've been able to minimise the number of 'extra' tags I've used.

3. With my mark-up finalised I can start full work on the stylesheet for screen (I might have skipped ahead and done some fonts and colours already, but nothing for layout). Positioning and tweaking the boxes comes first, then getting fonts and colours absolutely right along with any image replacement I want.

I'm intending to break down all future CSS I do into "layout" and "colours", for reasons coming below.

Finally, if appropriate, I'll add some "newer" CSS2 presentation that Internet Explorer might not like, but adds some artistic flourish.

4. Next I've hopefully finished my main 'screen' site, with the non-layout styles in a separate file. This allows me to apply just the colours and fonts to the stylesheets for handheld devices, avoiding blandness on those browsers.

5. Finally, I'll do a simple print stylesheet for completeness, though I don't tend to put too much into it. I'll hide non-content and increase the horizontal margins. I'll also add an ::after pseudo-element to links so I can display the URI in brackets after a link. Page numbering and so forth I'm happy to leave in the hands of the web browser.

And that, is how I make a web site. Or thereabouts.

 

[tRoG]

I like to do it, mainly because I find writing with XHTML + CSS just makes more sense and gives nicer code than having dodgy html tags all over the place, but I'd like to think that by sticking to the standards I'm contributing to a better web!

Also, by sticking to the standards, it means your page will be displayed more or less the same in all browsers (With a few very annoying exceptions). That's teh win.

So, yes, worth doing, but I wouldn't lose sleep over one or two minor errors.

 

[Maljonic]

Okay, can someone tell me what's wrong with this? http://validator.w3.org/check?uri=http://www.maljonicsdreams.com

 

[JingleBells]

If you look at line 165 it reads:
<img src=gif/amazon.gif alt="Amazon.com" border=0 align="right">

this should read:

<img src="gif/amazon.gif" alt="Amazon.com" border="0" align="right">

All attribute values should be enclosed in quotes.

 

[Maljonic]

Damn, bloody Dreamweaver's done that when I moved the image file!

 

[Maljonic]

ah, the joys of find and replace.