How much should I worry about the W3 Validator?

Hi, if I run my site through the w3 validator I have lots of errors:

http://validator.w3.org/check?uri=http%3A%2F%2Fwww.greyaliengames.com%2Findex.php&charset=%28detect+automatically%29&doctype=Inline

However, apart from the <td background> not being a valid attribute error (!weird!) all the rest are with code that I've pasted in from other sources such as google adsense, flash and reflexive affiliate links. It moans about varous tags in the pasted code, should I worry about it? Maybe it can be commented out but still work or something? Feel free to view my source if you need to.

I know I'm missing some Alt tags, but I'm not worried about them right now.

Also my mailto in the footer seems to cause it to flip out. weird.

Well I'm still a web noob so I guess it's to be expected.

Any help which will expand my feeble knowledge is appreciated thanks!

You have the 'Apache HTTP Server Version 1.3' online manual on your site.

Most interesting.

I have a BBS that doesn't fit the W3 standard at all, but it's fully grabbed by google/yahoo/baidu.

I think taht's a standard that regards to the W3C rule, but doesn't mean the page won't work (my BBS works o IE6/7, Opera, FireFox without problem, but it doesn't fit W3C).

Take it easy.

I see SSH server accepts Version 1.x connections. SSH protocol Version 1 has various vulnerabilties, this should be disabled and only version 2 clients should be allowed to connect.

This vulnerability check will be triggered when these conditions are met: TCP Port is OPEN "22" and VB Script returns TRUE (1) "openssh-ver1.vbs"

A few of those errors can be fixed by replacing the & with &amp;

And <embed> is a one sided tag so it should end like <embed /> not <embed></embed>. Not sure about the attributes for embed tho.

I think keeping to standards is a good thing and can help you in the long run in terms of future proofing, but I wouldn't be too worried, your errors certainly don't seem that severe.

No that's not wierd at all. See how there's no "background" attribute? The W3C validator does not validate Microsofts bastardization of HTML.

I would say yes. But you might not care that much about your customers.

So long as you've tested your site in all major browsers on multiple platforms, I can't see there being a problem. Sticking to the standards doesn't even mean that your site would render the same in different browsers (or even different versions of the same browser).

If you want to get all anal about it, get the browser booklet from VisiBone. It's an excellent resource.

regarding td background: It works in Safari, IE and Mozilla so even if it's a MS bastardization it still works...

So why bother validating in the first place?

Pretty much what Flamey said.

The only thing I don't bother validating 100% is my CSS code, since I intentionally have to break the rules with it using some hacks just to get my stupid pages working right in InternetExplorer.....


Regarding yer td background, why not just add a style tag for it to your CSS code if you need it? I'm a bit of a standards whore (aside from CSS for noted reasons), so I always check my pages for 100% compatibility (I use XHTML1.0 Strict), just to ensure I'm doing things the way the standards were defined, it's just unfortunate that a specific company and web-browser that doesn't need mentioning requires some extra tweaks to get said proper code working correctly.

No, that's what decss is for.

Wanna see a CSS nightmare? www.myspace.com/dampes8n

Look at the CSS for my myspace.. Talk about convoluted.

yeah I gather I could do the td in CSS. In fact I should really do all TD height and width in CSS too since they are "depreciated" now. I'd just like it to be "clean". Other simpler sites I've made (no addins) are...

If you want to be able to say "I don't care that it doesn't work in your standards-defying browser; this web site is standards compliant!!", then passing the validator is a good thing.
Otherwise, it's usually meh.

It is nice to be able to say you have a standard web site, and I for one will run web sites through the checker when I want to decide if I like something or not :)
(Especially content management systems, of course).
If they can't be bothered to follow nice web design practices, I'm not going to be paying much for their software.

Of course, since your stuff is nice, inexpensive casual stuff, the errors really aren't horrible and the web site looks nice, it isn't too much of a problem.

I find that getting the validator cooperating is often fairly easy, though, and worthwhile simply to feel good inside. Fixing one error magically repairs ten others. (A bit like compiler errors), and the errors are usually fixed rather quickly with alternative tags that do the same thing.

For example, I bet that background attribute could be replaced with style="background-image:url('pathtoimage');" and it would work.
From a design standpoint, I think CSS is a nicer thing to use than those arbitrary tags anyway, since it doesn't feel quite as close to just being dropped completely. (At least the arbitrary names for things in CSS are somewhat specific!).

Thanks for the feedback. What I wonder is how to allow the google, flash and reflexive AID links without the errors, any ideas?

Like this:

... arcade.reflexive.com/downloadgame.aspx?CID=22121&amp;AID=503"><b>Download ...

Usually I try to write valid sites, but if it starts to error on php links where it doesn't like & and = symbols in links I just leave it as it is, though it's just encoding type i should change:

http://www.f-once.pri.ee
http://www.e-hydepark.com/new/

dont i guess, every site ive put through it seems to have loads of errors including bbc, google and yahoo themselves (except msn interestingly enough :/).

whats the advantage of a site passing this?


I thought having good tags & robots.txt was the important bit:
http://www.widexl.com/remote/search-engines/metatag-analyzer.html

Thansk fredborg, I'll try it out.

I'd have to say that the main reason for using CSS and proper formatting is that it generally makes sites easier to maintain. The benefit is for the author, not so much the user (because browsers can handle extremely sloppy code, as has been noted).

I don't think I've ever checked the CSS or HTML of the site of any company I've bought from. So in commercial terms, assuming I'm the norm, it doesn't matter whether your pages validate. The worst that will happen is that more meticulous web authors will probably consider you a bit of an irritating moron for not taking the time to write proper code. And it is a bit pointless declaring a doctype if you're not going to stick to it. :D

I would suggest using the validator like compiler warnings, it will help you find actual errors and you get to learn how to code "correctly".

Regarding coding websites, I plan in my next little hobby project to make something as clean as Zen Garden, where everything in terms of layout and formatting is essentially handled by a linked CSS file in a nice, clean, fashion.

Not to mention easily modifiable as well, as to which is that site's aim, to allow artists to redesign the graphical appeal/layout of the site using the exact same page source, but their own CSS modifications.
(been trying to learn how to work cookies here so I could have multiple skins for users to select)

Another advantage of following standards is that your website may be more useful to people using special equipment, for instance things to help blind people read websites. On the other hand such technologies would probably be made to work with non-standard code as well.

Of course the biggest advantage is that you can put a banner on your site saying it passed the test!

There are a lot of reasons for ensuring your site validates with whatever doctype you are using.
Your site is likely to load faster, as the browser doesn't have to decide what to do with non-compliant code, or possibly even switch to a different mode of rendering just to display the site.
Your site is going to continue to work in the future, as browsers become less tolerant of non-compliant code.
Even moreso than the browsers, search engine spiders can index compliant code faster.
If you do contract work, a potential employer is likely to be judging your website before ever doing an interview.
Postings above list other reasons.
Plus, it is just a good feeling to know your site is valid.

Mark.

yes I heard about search engines favouring "correct" sites insome way or another.

Should you worry about it? not in the slightest. For commercial success the w3 validator means nothing.

google.com, yahoo.com, amazon.com, play.com, microsoft.com, symantec.com all fail the validator and all make a lot more money than anyone on this forum.

In the real world, it means less than giving your variables a meaningful name.

haha OK

That's nice. You realize that with the possible exception of play.com all the corporate websites you mentioned belong to companies that where commercially successful before there even was a world wide web, much less a standards governing body.

Yeah.. Google used to make napkins >.>

Yeah, and?