GDVS — Mon, 03 Apr 2006 11:46:14 +0000

I came across this at David Shea's site while reading up on sometheing else and thought some of you might be interested. It's about a completely different topic (using tables for layout) but the conclusion is the same, "Semantics only get you so far; at some point, you have to compromise and just use the tools you have at your disposal.".

GDVS — Sun, 02 Apr 2006 22:25:03 +0000

Busy wrote: If we had to wait until something was finished before using it 90% of people would be doomed.

My point exactly, the only real standard is what's implemented. SMB is a standard yet the Microsoft implementation extended it and changed all sorts of things. The Samba guys can't turn round and say "We will only follow the published standard" because it won't work with MS servers, they have to work with what's out there because that's the real standard not what's in an outdated document.

All the major browsers support XMLHttpRequest therefore it's a de facto standard, the fact the ECMA haven't got round to codifying it yet is rather beside the point.

Busy — Sun, 02 Apr 2006 21:18:34 +0000

If we had to wait until something was finished before using it 90% of people would be doomed.

I'm still waiting for Windows 95 to be finished, and 98, ME, 2000 ...
HTML is still not 100% supported in all browsers -each browser has a few tags they ignore or add

GDVS — Sun, 02 Apr 2006 17:43:38 +0000

Renegade wrote: Well, I don't think it's being extreme at all, what is the point of having web standards and validation if you're not going to follow it?

Define standards. Take CSS for example, the CSS 1 spec is a recommendation which in W3C speak means it's a finished product. CSS 2.0 was revised into CSS 2.1 because no-one was even close to implementing it succesfully and it remains a working draft to this day rather than a recommendation. Is that a standard? How about CSS 3, that's actively being worked on so is that a standard? Where do we draw the line? If you want to be ultra standards compliant then it's CSS 1.0 all the way, none of the other versions are officially done yet. Or you could bend with the wind and use the parts of 2.1 that are implemented while waiting for 3 to be finalised.

XMLHttpRequest is supported by all the major browsers whereas NONE of the CSS specifications are, not fully. Which is more of a standard? The one that's written down in the hopes people will one day comply or the one that's out there and working right now?

XMLHttpRequest will eventually be part of the ECMA spec, that's pretty much a given. Refusing to use it until then is just shooting yourself in the foot.

timjpriebe — Mon, 27 Mar 2006 14:03:13 +0000

Complying to standards in the web world can be compared to complying with standards in the building world, although it's not a perfect parallel.

Are the rules there for a reason?

Yes.

Would we be better off if everyone complied with those rules?

Yes.

Is that going to happen?

No.

Is 100% perfect compliance really necessary in most cases?

No.

Would it hurt anything to be 100% perfectly compliance?

Aside from time spent complying, no.

Would I fault anyone for not wanting to use something that's 100% compliant?

No.

Would I fault anyone for using it if it's not 100% compliant?

No.

If I think something should be 100% compliant, am I going to take the earliest opportunity to point that out to the owner?

No, that will very likely turn them off. If it's so non-compliant that it just doesn't work, then I probably will. If they're having major code problems, then sure, I'll tell them to comply first and see if that doesn't fix their problems. But if they ask for a critique for their site, I'm not going to start off forcing compliance on them.

If anyone else advocates compliance and does go with the tactic of mentioning compliance first, should I jump down their throat?

In the same vein, that wouldn't get them to change. If they choose to critique something by first mentioning compliance, that's their choice.

porto88 — Sat, 25 Mar 2006 21:51:26 +0000

teammatt3 wrote: Don't you use Gmail? Isn't that Ajax? Google search pages don't validate, do we boycott Google? I think not.

I know I will go to Code Hell for saying this but I don't give a crap about valid code. I see see no benefit from valid code. Anyway, I'm getting off topic here so back to Writely. Pretend it validates, it's a pretty cool app now huh?

i agree completely, you guys are a all little babies

teammatt3 — Mon, 13 Mar 2006 02:42:10 +0000

Don't you use Gmail? Isn't that Ajax? Google search pages don't validate, do we boycott Google? I think not.

I know I will go to Code Hell for saying this but I don't give a crap about valid code. I see see no benefit from valid code. Anyway, I'm getting off topic here so back to Writely. Pretend it validates, it's a pretty cool app now huh?

Renegade — Mon, 13 Mar 2006 02:28:29 +0000

Well, I don't think it's being extreme at all, what is the point of having web standards and validation if you're not going to follow it? AJAX depends on that one non standard function so, even though it may be "Web 2.0" the supposed "future" of the web, we're actually going backwards by using non standard coding practices.

Roo — Sun, 12 Mar 2006 17:58:25 +0000

Well with so many out there that are valid, or can be made valid with a bit of tweakage....yeah...I'd refuse to use it because it isn't valid.

As a matter of fact I need a dropdown menu and I'm refusing to use dhtml scripts I'm finding because they don't validate. I'm trying to get ALA's 'Suckerfish driopdowns' done with CSS to work in IE...

teammatt3 — Sun, 12 Mar 2006 16:05:44 +0000

You guys refuse to use web applications just because it has non-valid code? You gotta be kidding me. I know having valid code is important to you but that's taking it to the extreme.