timjpriebe — Tue, 17 Oct 2006 12:47:38 +0000

This is not a problem for me personally. I like to validate when possible, but when validation and business concerns clash, even when not in direct conflict, the business concerns win out every time.

Megan — Mon, 16 Oct 2006 13:31:29 +0000

Ha! Read that more closely this time. Most of it's good, except for the "complex and useless rules" which I definitely strongly disagree with!

Busy — Fri, 13 Oct 2006 21:18:50 +0000

I like the validator on the digg site (click on the articles title on the page Galeenet posted)
be sure to read the select box options on the validator page

Galeenet — Fri, 13 Oct 2006 15:36:51 +0000

Nice to see I'm not the only one getting tired of relentless W3C obsession.

I found the thread by http://digg.com/design/Who_cares_about_XHTML_validation_anyway

About validation, I found it mandatory for content meaned to be integrated elsewhere (thirds parties, affiliates). In other cases, coding along the standards is good, but finding 'W3C stickers' on some e-com sites is just laughable.

(pardon my english, I'm french ! )

Megan — Fri, 13 Oct 2006 15:20:40 +0000

demonhale;207904 wrote: True, if you have time, you can actually look into the php scripts and edit the output to validate but is very time consuming... .

I just came back to this thread because I wanted to link to it from something else. Didn't see these responses before.

Mark and I think Liam (I think - I know Mark did) both looked at that phpAdsNew code and couldn't find an easy way to fix it. Busy, I'll have Liam PM you with details. It's got something to do with how phpAdsNew generates the code.

demonhale — Sat, 30 Sep 2006 03:34:37 +0000

Busy wrote: I think the PHP excuses given are just that - excuses.

True, if you have time, you can actually look into the php scripts and edit the output to validate but is very time consuming...

I did it once to a feedback board script and sent the validated output of the source back to the original author of the script, now all I do is fix my own installations if I had "time"

Megan wrote: But I'm a full time professional and I have time to worry about these things. And if I want to be great at what I do, I should worry about these things

This could be my argument too, totally agree with this...

It actually helps to validate, cross-browser issues are solved this way...

Busy — Fri, 29 Sep 2006 20:44:59 +0000

Megan;207873 wrote: Do you want to fix the phpAds new code then? Or is that not what you're referring to?

Sure, I'm busy for next few days though as I need to catch up (been away for 3 days), end of next week can have a look at it for you if you want, PM me the details.

Megan — Fri, 29 Sep 2006 13:03:55 +0000

Busy;207863 wrote: I think the PHP excuses given are just that - excuses. .

Do you want to fix the phpAds new code then? Or is that not what you're referring to?

I definitely think there is a difference between web standards and validation - validation is part of web standards but certainly not all of it. I'm not sure if the author of the original statement was making that distinction or not. So I'll break this down into two parts:

Is it possible to be too obsessed with validation? Yes, if it means not being aware of other aspects of web standards (i.e. semantics). It's like running a spell check on a written document and not proofreading it yourself. It doesn't catch everything - ever see those poems English teachers will post to demonstrate why spell check's don't catch everything? Vaildation is like that.

For me, I'll admit that not seeing that green bar irks me a bit. There are 136 errors on my personal site right now because of & and = signs in urls I've copied and pasted. I could go back and change all of them. And I admit, I'm tempted, but it's not exactly a great use of my time right now. We've also got a bunch of errors on this site because of the phpAds New code which I mentioned before. There's not an easy way to fix that but we plan to do it eventually.

Is it possible to be too obsessed with web standards? I think it is, to be honest. So many standards experts seem to be completely oblivious and uninterested in other web design problems like marketing and SEO. It's important to take a balanced approach: web standards are great, and everyone should follow them, but they're not everything or even IMO the most important aspect of a good website.

(I'd also like to make a big distinction between writing really sloppy 90's era code and something that is almost standards compliant. 90's era code: bad, should be elimitated. Almost standards compliant: well, it's not going to kill you )

For me, I admit that I sometimes obsess over the most correct way of doing something. Like what's the exact correct markup for something, or the best name for an ID. Or whether I should evn be using an ID or if I can use the cascade to style that element instead. But I'm a full time professional and I have time to worry about these things. And if I want to be great at what I do, I should worry about these things

Busy — Fri, 29 Sep 2006 09:08:05 +0000

I like to validate just for piece of mind. Although I only usually get it validated on build and don't worry much after updating. I don't worry about CSS warnings, only errors.

I think the PHP excuses given are just that - excuses.
Just about all (x)html can be valid whether it be server side or not, if a server side page/script doesn't validate it usually comes down to bad coding. The biggest problem is browser issues, but that usually resides within CSS so at least the (x)html can be valid even if the CSS isn't due to hacks or whatever.

Abhishek Reddy — Fri, 29 Sep 2006 03:20:39 +0000

No. Well, sometimes. If I can make it pass the validation test within the same order of effort/time that it took to set it up to mostly-valid but functioning standard, then I will.

demonhale raises an interesting point, that XHTML code is often generated by underlying programs. I find web scripts and programs are frequently poorly designed or architected, and end up producing invalid code because of it ("scripting conflicts" shouldn't happen). Fixing it is difficult because the cause of the problem lies in a PHP (or whatever) soup, often asking for a rewrite. I too leave things be, in such a case.

It's perfectly okay for a standards aware developer to be less than obsessed with validating, at present. As technology evolves, I think it will become easier and more necessary to create fully valid code -- as we drift towards XML, well-formedness both for human and machine readability becomes important.