How do I answer the question "why validate" Options

[Sharondippity]

I provided this link http://validator.w3.org/docs/why.html

and the person insists that validating is silly. This person is a webmaster.

Where else can I point this person for proof?
I'm flummoxed.


:?:

[bragadocchio]

Hi Sharon,

There are some people who refuse to ask for directions when they know they are lost. I do that sometimes myself. It's usually a mistake. :oops:

I validate because I make mistakes. I sometimes leave out something important like a closing tag somewhere, or I spell something wrong, or something equally stupid.

I validate because I am a little bit lazy. A validator finds the mistakes a lot easier than I do. Life's too short spending more than a couple of minutes looking for mistakes you might have made when a validator can find them faster and easier.

This is a good blog post on the subject, from Mark Pilgrim:

Why we won't help you

[Sharondippity]

Thank you, that puts a nice spin on it.

[bragadocchio]

You're welcome.

I'm not afraid to admit that I'm willing to follow standards because they make it easier to avoid pages that have mistakes on them, or don't show up right in some browsers.

And a validation program doesn't pick on you when you make a silly mistake.

[Sharondippity]

LOL!

NOw where's the fun in that I ask..

[bragadocchio]

It gives you time to have fun doing other stuff.

Things that are more fun than grumbling about how your page doesn't work right, and looks like garbage in four out of five browsers, but good in Internet Explorer

[sanity]

Was the person interested in the topic initially or did you bring it up?

[sonjay]

I can only guess that your friend wouldn't mind if his mechanic worked on his car without making sure his tools were properly calibrated. You know how it frequently says "Tighten nuts to xx amount of torque"? Bah, who needs that?

He probably doesn't care if the plumber makes sure that the pipe fittings are the right size, either.

It probably doesn't bother him if the grocery store where he shops keeps their refrigerators and freezers maintained to meet the standards for refrigerated and frozen foods.

And I'm quite sure it doesn't bother him at all when his computer crashes because a software programmer didn't debug his code.

Heck, he probably gets to feeling all warm and fuzzy when his surgeon says "Oh, I don't worry about how your surgery is supposed to be done; the prescribed methods for this procedure get in the way of my creativity. I just hack away in there until things look right to me."

[Sharondippity]

He was adding to someone else's topic.

A responsible webmaster will test a site with whatever browsers are considered to be important (which statistically is just IE). Microsoft's FrontPage produces script that runs in its browser prefectly. Whether or not it validates elsewhere is irrelevant. 

The badge is silly. It does nothing for the visitor, or customer, so is meaningless. Ther is no connection between not crashing a browser and validating.

[bragadocchio]

By the "badge", I'm assuming that he was referring to the button that you can put on a page to show that a page validates. I'll agree that it is silly. It doesn't do anything for the visitor or the customer.

But there's nothing that requires a person to use that button if they use a validator.

And, writing code that might fail for 1 in 10 of a site's visitors, or more, is irresponsible if a client expects the site to work.

There have been more than a million downloads a day of Firefox since it went out of Beta on November 9th. It's faster, smaller, more secure, and has greater functionality that Internet Explorer. The 1 in 10 I mentioned above will probably become 1 in 5 in not too much time, and Microsoft's lack of competitiveness in adding to Internet Explorer will shrink their dominance of the market even more.

Even the programmer writer for Frontpage suggests that people test in other browsers:

http://blogs.msdn.com/lisawoll/archive/200.../04/237786.aspx

[Adrian]

whatever browsers are considered to be important (which statistically is just IE)

Uh huh, and there's already discussion about IE losing market share, which seems set to continue on the same trend for a little while at least, with all the publicity FireFox has had jsut recently, and Mozilla and Opera proving popular choices in areas.

So though 6 months ago your site that only works in IE was fine in 95% of browsers, now it's down to around 90% accoridng to popular stats, and is set to drop some more.

Is the relatively small amount of effort required to make a site validate better, too much to think about, as opposed to going abck and rewriting a site because the customer suddenly finds 20% of their business can't use a site?

I'm sat here spending thousands of pounds a year on computer equipment, and I use Opera. If a suppliers site doesn't work in Opera, they have to give me a lot more reason to continue using them over a competitors whos does work in Opera.

If your customer is general Joe Public, IE is probably a safer bet, but if you're working in any kind of technology sector, I'd imagine the percentages of non-IE browsers was a bit higher.

Validation isn't silly. My response to that would be, why is it silly? And don't quote the silly thing about IE being the only important browser.

[projectphp]

Microsoft's FrontPage produces script that runs in its browser prefectly. Whether or not it validates elsewhere is irrelevant.

Except that what works for IE may not be Search Engine friendly...

I had a site once that worked AOK. Thing was, they didn't put their links in qoutation marks: e.g. <a href=http://somepage.com> Inktomi had a fit.

Another time, the site put frameset declarations live with end of line characters. Fine for IE, bad for SE.

Anyone that says that producing pages for IE is all that is required is probably the same person that has no idea that marketing on the web is more than a case of "build it and they will come".

Validating takes one second, and often finds damaging problems. The question IMHO isn't why validate, but why not?

[DianeV]

Or, as the fellow who helped write the Miva programming language said, they're coding for Microsoft, not the Web.

By the way, in some industries, the percentage of people using IE is only 75% now: http://www.w3schools.com/browsers/browsers_stats.asp

Given that Opera can also be set to identify itself as IE (and users like me use that setting due to designers who code Microsoft-only websites), that number may be higher.

[gravelsack]

The question IMHO isn't why validate, but why not?

Because I can't be bothered.
Because I don't do much hand coding anyway and am happy with the standard non-validation issues produced by Dreamweaver.
Because my sites work perfectly well even with loads of reported errors.
Because the time I might use to fix the 'validation' errors could be better used to make more pages (that don't validate for the same reasons) and make more money.
Because some of the vaildation errors are part of my own personal SEO techniques.
Because nobody has yet been able to show me the financial benefit in so doing.
Because I am not anally-retentive.

.... probably more as well.


I am not against other people validating in the same way I am not against people wearing suits and ties - just not needed for me, thanks. When it becomes essential, I'll validate like hell.

[projectphp]

When it becomes essential, I'll validate like hell.

When that day arrives, it may be too late...

Because nobody has yet been able to show me the financial benefit in so doing.

There are two issues here: absolute validation and just checking. Absolute validation has no benefit. Zero. In fact, it may have a negative ROI. But just checking and fixing major flaws? That has a huge benefit.

IMHO, so much can and will go wrong, that the more one tries to build sites that follow a standard, the better they will work. With MSN launching its own cralwer, chances are many IE coding non-errors may cause MSNBot errors. Crawlers start primitive and get better. Why cahnce it when 20-40% of the SE market, and a higher percentage of unique users, may be lost?

Not a financial benefit in the short term, but longer term well worth it IMHO.

[DianeV]

Good, Michael, for making a distinction between absolute validation and checking code.

I'm mainly talking about fixing broken code or incorrect CSS syntax; the W3 validators are good for that.

I use Dreamweaver often, and I should note that on rare occasions, when dragging text and links around the page, DW has left a copy of the <a href="whatever"></a> in the original spot -- so now there's an inadvertent hidden link.

OTOH, I like the visual display of & 151 ; rather than the double dash (--). The validator doesn't like it. Oh well.

[Adrian]

First thing I tend to suggest people when they have a rendering problem is that they run it through the validator and clean up what they can.

I've seen many people solve rendering issues that way before....

Anally retentive? really? You think thats what its about?
Personally, I feel if you think thats the case, all you've seen is some hype and no proper information on it.

100% validation is rarely acheived, and just as rarely really truely sought after.
Certain validation proecesses can improve things like SEO and usability and accessibility.

But if you knowingly do something different, fair enough. Just like if you knowingly use an SEO technique known to get sites banned.....

Because I can't be bothered. 
Because I don't do much hand coding anyway and am happy with the standard non-validation issues produced by Dreamweaver. 
Because my sites work perfectly well even with loads of reported errors. 

Then be prepared that someone who can be bothered, who isn't happy to put up with badly generated WYSIWYG code, whos site works well in several browsers now, and in future releases, is going to get some business that you're going to miss out on. or is going to have time later on, to make more pages, to make more money, while you're recoding the whole site to take account of IE7 bugs.

[gravelsack]

When that day arrives, it may be too late...

Don't care

I could fix most of my sites retrospectively in pretty quick order. Heck, almost all of them validate already (in the non-absolute sense).

But just checking and fixing major flaws? That has a huge benefit.

Agreed - but as far as I am concernded if the site displays OK in IE, Mozilla, Opera and my mate's Mac stuff, then there are no major flaws.

The point I am making is that I check by practical definitions of 'quality' not theoretical. My decisions are ROI driven. If I produce a template that doesn't work in one of the browsers, and I can't immediately see why, I'll put it through a validator - otherwise, why worry?

Its far better to develop a style of working that has quality built in than to rely on quality control at the end of the process.

Then be prepared that someone who can be bothered, who isn't happy to put up with badly generated WYSIWYG code, whos site works well in several browsers now, and in future releases, is going to get some business that you're going to miss out on. or is going to have time later on, to make more pages, to make more money, while you're recoding the whole site to take account of IE7 bugs.

Adrian, if you really believe this then you do not know how to use Dreamweaver properly. We have been round the houses on this in another thread, and I thought I had pretty much killed this nonsense off - PM'd proof to someone anyway.

You are making assumptions and then jumping to incorrect conclusions as a result. Shoddy, old chap - maybe a thought process validator would have helped (that's a BTW)

You will not be able to cut any code from my Dreamweaver stuff - not a jot. Everything is driven by external css and the html is exactly as slick as any of your hand coded html - probably slicker than most people's.

No font tags, no table tags (unless needed), no positioning tags - just <div id="divname">, some redefined html tags in css, and the raw content.

All done by Dreamweaver and one free extension.

The most that usually needs doing by hand is a bit of tweaking to the css to keep it simple.

If you get the quality right at the start, the site will validate well enough. Anyone still doing the old fashioned 'slow coding and restrospective quality control' stuff is losing business to me.

(added)
I just checked my last three sites - all have far less errors than the forum site we are now viewing

[Adrian]

then you're doing a lot more than most WYSIWYG users Gravelsack, and doing what a lot of actual standards advocates do do anyway

Validate to a practical sense, and leave in errors in full knowledge of the issues at hand.

My comments were more for people who hardly take any heed of web standards and validation, rather than those who do a fair bit, but not 100%

[gravelsack]

No offense taken (or intended in the opposite direction) - I just like a rant every now and then.

DW is just a tool - and its one that I will need to leave behind soon as some of the css I want to do doesn't render correctly (and what use is WYSIWYG, if it doesn't WYS)

I'll probably switch to NVU, if it keeps improving - will I validate then? You bet your A** I will - at least until I know what it is doing (and what I am doing, for that matter;)).