How many of you guys use valid HTML/XHTML?

Oh, how inconsiderate of me. "drhowarddrfine," has already settled this discussion years ago, and therefore there is no need to bring it up in discussion.

Did the thought ever occur to you that you were not the center of the universe? That maybe there were other people living in this world with unresolved issues?

Where did I state this? Where did I even hint at the fact that this question is completely unique, and never asked before? This way of thinking comes from people like you who are always looking to grab the intellectual spotlight in any kind of trivial situation.

I'm done talking with you, your a mocker and I (along with many other people on this forum) do not approve of the way you word things. I am sure you have learned this by now, but your words will hold you accountable and you can expect a rebuttal, when you say the things that you do, on a forum such as this. Yeah, I'm 19 years old, but I also work 8+ hours a day on my internet business, while being a full-time college student, and working for a local newspaper. Not everyone of my piers has chosen the same lifestyle, but you won't see me flaunting about it or mocking them. I take my business seriously, regardless of whether or not there are seasoned veterans with 40+ years experience in the field. Do I seek their advice? Yes, but not from people like you.

This is the kind of discussion I was looking for. No self-righteous emotional rants, mmerlinn is a professional (or talks like one) and is shedding his advice based on his past experience. This is why I started this thread, so I could read constructive things like this, and form my own opinion based on collective reasoning.

 

*edit: Scott, my real answer to you is at the bottom, you posted before I did, so... everything in the middle is just bleh : )

So does XML. The problem here is there's been a long-going debate over whether HTML (or especially XHTML if it's actually sent as that mime-type, or how about RSS aggregators like Atom?) should be that draconian. On the one hand, having every single browser stop at the first error, throw up its hands and say "There's an error on line X and this page will not render until it's fixed" and sit there like an obstinate child may well make the web a better place in that, you'd think the author will then fix the error.

The number of aging pages who don't have maintaining authors with useful information pretty much speaks against that, though. And as the link Doc first linked to (Pilgrim's rant, a copy of what Ovid and others over at PM have been saying since forever, and a nice sidebar to what Eric Raymond posted so long ago) shows the problem with this (and, I should mention Pilgrim is the one who was against Atom treating XML like... XML, instead of taking shit like RSS2 does...) where there are many sites where the author does not have the control (comments, ampersands, yay.).

Me, I sure would want my browsers to at least initially tell me the page has errors, at least while I'm building. After it's released into the wild I have little control over it (it means I have to nag at my colleague and now that he has a girlfriend he's got the nagging::ignore module working pretty well : ) and so...

this is why I validate. It's the only time I can see my errors, BEFORE the page is released Out There, the only time I have someone to check my spelling and make sure I've dotted my i's and crossed my t's.

Now I'll go through Eric's comments : )

I think you guys are talking past each other. It sounds like you both know that validation can remove errors that cause browsers to puke, and yet at the same time a page can be perfectly valid and still not cross-browser compatible due to some vendor being an ass or missing a bug.

He cares. Maybe too much, but he cares about validation. And yes, missing an alt attribute is even something I've done when I'm usually pretty conscious about them.

Actually, you can (many of us do) hack around IE (and anyone else) and make it all valid. That aside, there are things that many of us will still go ahead and stick in our CSS that aren't valid (like behavior to include PeterNed's whatever:hover for IE6, for instance) because it's an invalidity we're willing to deal with. There are valid ways to feed that to IE but people like me like to keep our dirty hacks in the single CSS sheet rather than using constant IE Conditional Comments on every HTML page (which are valid, if evil).

True enough. I've got a display: inline-block menu on a site right now and yep, it's an error as far as the validator's concerned because it's programmed to flag anything it doesn't recognise as an error. In the community, though, it's considered ok as it's not some acccidental thing but a vendor testing code (what Mozilla uses to test properties before they fully release them as actual features). Every vendor has them. If Doc's saying they're not invalid, he must mean they are not considered an invalid use in the community. The validator's a spellchecker. It doesn't know some words. My spellchecker right here right now in FF doesn't recognise the word "validator". But nobody would tell me it's invalid for me to use it. Rather, the machine simply doesn't know it (and I like it that way, actually-- I want vendor tags to be flagged as errors).

You mean, because they don't work in IE? Bah. Let IE users miss out on some of the good stuff (while having full acccessibility to content and the page should work fully for them) while everyone else gets the better experience.

You guys are talking past each other again. I think Doc thought you were saying CSS3 things were invalid because the CSS2.1 validator says so, even though you both know how this all works. This actually relates to the issue above re -moz etc.

I fully agree. There are some good things in HTML, the best being the "every browser reacting the same way to errors" thing, and the shorter doctype... but there are a lot of things in there (I heard Anne van Kesteren present a bunch of these just a week or two ago) that either belong in CSS (like setting colours and widths for things... that does NOT belong in a markup language... let's not go back to <embed width="blah" type stuff again) or are the job of the back-end/server (some of the functions the new tags do really shouldn't be running on the client AT ALL, even though newer browsers and newer computers get more and more powerful... not all of them are, and the information super highway should work for everyone!). While I may get lazy enough to use that awesome doctype, I doubt I will ever lower myself to using many of the tags. They are mingling content and presentation again (though one could argue with some applications the line between them is pretty blurry).

Bullets are in your HTML? They are decorative-- so why aren't they in the CSS as background images, where they belong?
But if you can't for some reason, something meaningless like a bullet should have this:
alt=""
which means, there is no alternative meaning, because there is NO meaning.

For small buttons, if they are clickable or do anything, they need alt text. Otherwise, they get what HTML bullets should get: alt=""

: )

To the OP:
If you are having a lot of trouble validating and it's becoming a pain, I can tell you from experience that you either a) have some nastiness in your own code, or b) have nastiness in someone else's code, like you get in widgets and Google ads etc.

I know that validating well-written and semantic code is pretty easy, and when you're good you won't have hardly any errors the first time you go the validator (hint, don't wait til large chunks of the page are all finished-- validate as each chunk is finished). I also know the kind of code I first put out when I was starting out was rife with errors and I had a hard time figuring out what exactly was so wrong in the first place. Also, at the time the validator didn't give those handy little explanations at the bottom. I remember the first time my husband was told his <head> wasn't closed, while we stared at the big </head> sitting right there... nowadays the validator says <head>s considered not closed if you don't have the required minimum child (<title>) in it. That sort of thing.

I validate, every time I make a major change (and sometimes after a minor one) and right before I think a page is ready. I validate constantly, knowing that it will not stop browser bugs but that it is a great tool which assists me in web building... just like the spell checker in the text editor you are writing your CV/resume in is a very very handy tool. Not using it, not trying to correct your errors because they are overwhelming you, means your document is certainly not ready yet. You may need some help-- that's fine. Ask. Nobody wants crap code floating around Out There, and you obviously care about your code or you wouldn't be trying to validate in the first place. So, we're here. Even Doc, in his crusty way. Please keep validating and trying to get it right! Not "just because"-- there's a reason for the madness.
I think that was what you were asking-- why am I validating, and how much does it really matter.

 

Yea we need to report his posts and get this guy banned

Yes sir you are correct. But notice my exact words. I said "sometimes don't validate.". Meaning that it would depend on the hack in question on whether it would validate

While I am all for letting IE users suffer, sadly I am one of them. In school I have to use the Ie that is on the computers (7). However I got tired of IE and decided to download firefox on my flash drive. I currently browser with firefox, breaking school rules because I am tired of IE. My teacher tried to get me in trouble with using Firefox so I threw a bunch of complicated words and why IE is bad for like 5 minutes and he was just like..oh..ok then... [how in the heck did i get started talking about this???]

Yes and Tommy Olsen does too. They are heading the wrong way with html 5. Hopefully they will fix many of these errors. Truth be told, I am perfectly happy with the standards right now (except css, I cant wait for css3, I am more talking of html standard). The main thing we need now are more compatible browsers, or have IE die off.

~eric

ps. When i say compatible browsers I only mean IE. Not firefox or Opera or Chrome etc etc. I found a promising web browser called "Flock" (google it) and its on the same engine as Firefox. It is quite nice actually. I am interested to see its future

 

If you want a free, small, super fast, fully standards compliant Geko (Mozilla) based Browser for windows try K-Meleon (play on Chameleon, get it?). The desktop icon tself is worth the effort to download it.

James

 

Most pages that show up in search results do not validate. However, it probably is a good idea to learn proper coding.

 

Well yea. Strive to code valid html 4.01 (or whatever doctype you prefer). But don't get all mad and waste time if you don't/can't have valid code without changing your site. Just try and code by the standards. I think this thread is pretty much cleared up.

~eric

 

In my opinion I don't think the validity of a page and the presentation of a page have anything in common, therefore whilst I agree that most "famous" websites are hugely invalid, and valid sites tend too have a bland appeal, there is no actual reason for this.

Any desired presentation of a website can be coded validly, even your most extravagent designs. So it is wrong to think that valid websites are boring because it doesn't make much sense.

For example embedding an swf is still the same SWF when embedded correctly or incorrectly, it doesn't take any appeal away from the SWF.

By not making your pages valid you are only decreasing it's ability to work correctly amongst different browsers, it isn't going to spice up your page..

 

Only that you never know how any particular browser will handle invalid code... as they each have their own different error rendering...

I think it's because those who can write valid code (and care enough to do so) are coders and not working with designers. Designers work where the money is good-- large companies with bad nerds who don't know how to write properly, or don't care, or don't have the time, or need to work with legacy code (ew).

Lawlz, it's a technical document. You know, specs. Like your car manual. It might list all the rules for how your wildly designed fancy automobile should be maintained. But the manual is still plain on paper.

*edit my earlier post has a bogosity... should be

Lawlz, HTML5 has the same-error-rendering thing, not HTML. I did mean 5, I swear. : )

 

Well, I did just that, today. I got it in Spanish for some reason : ) It's working on my Virtual box just fine now and yup, it's damn fast. It won't be replacing FF2 on this VB (as I use the Web Dev Toolbar) but it surfs great.

I expected the Chameleon you see on the site, not the goofy-looking dinosaur : )