I am a stickler for standards. Whether they’re coding standards, sporting standards, laboratory standards or, more relevant to the subject of this article, web standards. Standards enable uniformity and compatibility and essentially, they make life easier. Admittedly, I am all for anything that makes life easier provided it doesn’t breach my moral code and the laws of the land in which I reside. This is why it grinds my gears that web standards are treated with so little regard.
Earlier this week, while validating a website being launched for one of my clients with the W3C Validator Service, I decided to run a few other well known websites through the validator as well. All I really wanted to see was what doctypes were being used by some of the largest websites on the web. What I found was rather astonishing, if not disappointing and utterly upsetting. Have a look at the following table:
| WEBSITE | DOCTYPE | RESULTS |
| Google.com | HTML 5 | 39 Errors, 2 Warnings |
| Yahoo.com | HTML 4.01 Transitional | 34 Errors, 8 Warnings |
| MSN.com | XHTML 1.0 Strict | 1 Error |
| Amazon.com | HTML 4.01 Transitional | 1160 Errors, 79 Warnings |
| Facebook.com | XHTML 1.0 Strict | 44 Errors, 4 Warnings |
| Twitter.com | XHTML 1.0 Strict | 96 Errors |
| Mashable.com | XHTML 1.0 Transitional | 167 Errors, 44 Warnings |
| TechCrunch.com | XHTML 1.0 Strict | 200 Errors, 26 Warnings |
| Youtube.com | HTML 4.01 Transitional | 241 Errors, 60 Warnings |
| Microsoft.com | XHTML 1.0 Transitional | 103 Errors, 25 Warnings |
| Bing.com | XHTML 1.0 Transitional | 12 Errors |
| SitePoint.com | XHTML 1.0 Strict | 5 Errors, 5Warnings |
| WebStandards.org | XHTML 1.0 Strict | 1 Error |
Thirteen(13) of the largest, most popular and most visited websites in the world serve us webpages containing invalid markup. The Amazon.com homepage alone contains 1160 errors, more than the rest of the list combined. Even more egregious is the fact that the homepage of The Web Standards Project (WebStandards.org) contains an error and, as such, is not a valid XHTML document (#epicfail). Also of concern is that both Yahoo.com and Youtube.com, both websites in the top 10 most visited websites in the world, are using HTML 4.01 markup which on the 24th of December this year will be a 10 year old specification.
I figured that there must be standards compliant websites on the web and after a lot of searching (too much) I found only four:
Very disappointing…
This is one of the main reasons why very few web designers respect web standards. There is hardly any precedent. The influential leaders on the internet are not leading by example. One of the most hypocritical things I have ever seen in my life is how some of these companies and web designers go on about how bad Internet Explorer 6 is yet they themselves build websites choc full of invalid webpages. Internet Explorer 6 really is a bloated, overused wad of epic fail. However, with 241 errors in markup, Youtube.com (whose developers have recently declared that it will no longer support the craptastic IE6) is also a grand, epic fail. As a matter of fact all of the 13 websites I listed above fail. Very, very disappointing…
Another group of entities responsible for the lack of standards compliance on the web is the manufacturers of the web browsers we use. The Microsoft, the Mozilla, the Opera, the Apple, the Google and all the others. They don’t adhere to web standards either. The wide range of possibilities that exist with regards to how webpages are rendered in these browsers is just ridiculous. In my opinion there should only be one standards compliant mode for rendering web pages. Can you imagine the turmoil that would ensue if the FCC didn’t enforce radio transmission standards and regulations?
Additionally, the concept of having a ‘quirks mode’ in these browsers is one that has always rubbed me the wrong way. How do you enforce a web standard when the consequence of not adhering to said standard is simply that invalid pages will be rendered in quirks mode? The purpose of quirks mode is ‘to aid in the maintenance of backward compatibility with webpages designed for older browsers’. The fundamental flaw with that definition and by extension the concept of quirks mode itself is that webpages should not be designed for browsers, they should be designed to meet predefined standards!
The path that we’re on as web designers, browser manufacturers and internet businesses is one of self destruction. We’re making things harder for ourselves. When we decide to not make our websites and our web browsers standards compliant, we have to work harder to ensure that our pages display correctly across browsers, we have to build more bloated browser applications to ensure that our browsers can handle non-standard markup and we waste time arguing over stupid things. I want a standard web. A standard web would make my life and my work a whole lot easier. Web people, don’t you want your work to be easier? Don’t you want a standard web?
P.S. I am acutely aware that some pages on this website contain invalid markup. My frustration with some of the widget and plugin developers for WordPress is one of the reasons why I wrote this post.
Short URL for this post:
http://bit.ly/34TaS9
Liked this post? Disliked it? Let people know! Share it!
Corve from Saint Andrew, Jamaica
kjordan from Kansas, United States on 
{ 3 trackbacks }
{ 14 comments… read them below or add one }
Twitter: @pixelflips
Great post, although I have to disagree with a few points. While I do feel that valid markup could be better supported on the top sites you mention, I don’t think invalid code means these sites aren’t supporting web standards.
They are using correct doc types, using CSS for separating presentation from the content , and avoiding using tables for layouts. Which already shows a great deal of support for standards based design practices.
I don’t feel that validation should be the only factor when it comes to saying if a site is supporting web standards or not. This is only one of the factors to be considered in my opinion and is of course something we as designers should try and adhere to. Plus if validation is how we decide whether or not a site is supporting standards, then you aren’t supporting them either as this site has a few errors too!
Just a few of my thoughts. Keep up the great work and have a great weekend.
Cheers.
>> webpages should not be designed for browsers, they should be designed to meet predefined standards!
I disagree on both counts, webpages should be designed with the user in mind. If omitting some ‘required’ tags can allow a smaller page (or whatever it may allow) , and a better average user experience for all of the site’s users without any loss in content/interface, who cares about the standard? The goal is for the standard to serve the implementers and users, not the other way around.
You seem to be forgetting the most important aspect, the price.
A simple when served hundreds of millions of times ends up being Mbs of bandwidth. If it’s possible to miss out a few tags while having the same rendering and save thousands of dollars per year in delivery costs, wouldn’t you do it too ?
we all know why the google search page doesn’t adhere to the standards: every byte counts. they’re a special case.
the others? not so much. i agree they should try harder.
um … one thing i find interesting: try running this page through the validator: 33 Errors, 11 warning(s)
http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.samuelfolkes.com%2F2009%2F10%2Fweb-standards-compliance-or-the-lack-thereof%2F
33 Errors, 11 warning(s) of your own blog
Twitter: @SamuelFolkes
@Phillip – While I agree that markup validation isn’t the only factor to be considered I strongly feel that it is one of the key factors. I am aware that some pages on my site are not valid and frustration with some of the plugin authors for WordPress was one of the motivating factors for writing this post. The digg, dzone and retweet widgets are the culprits and it annoys me a great deal.
@Falaina – The requirements for valid xhtml markup are very few. I really doubt that ‘omitting some required tags’ would significantly decrease page size. I agree that the goal should be user-friendliness, however there must be some guidelines. Ralph Lauren cannot make jeans with sub-standard stitching because it makes the jeans lighter or more flexible.
@Eoin – Breaking a standard is not the way to go about cutting costs. That gets financial corporations in trouble with the SEC.
@schnalle & @Anonymous – See first response @Phillip
the validator often shows up errors for various minor reasons. After validating a site once people rarely go back and validate the site again.
It’s a shame that you spoil an otherwise good post with this. The age of the spec is irrelevant, particularly given that XHTML 1.0 (which you seem to prefer) offers precisely zero improvements over HTML 4.01. It’s simply a different way of expressing exactly the same thing, and one that adds unnecessary bloat for no gain. What’s perhaps of more concern is that there hasn’t been a new spec for 10 years. HTML 5, when it finally arrives, should help there. But in the mean time, HTML 4.01 is probably the best option to use now.
Hey Sam!
Respectable article, but the one thing I’d say is that the FCC sure had an easier time setting a standard when there were at most 50 radio stations (if so many). Like most things in this world, development of pages with valid Markup is driven by cost. How much time does it take for a youtube.com developer to produce perfectly valid markup and what additional benefits does this bring and how do these benefits match up to the time and monetary costs??
You forgot to mention Wikipedia, which is ”one of the largest, most popular…” .
Wikipedia is a semantic and markup model.
Twitter: @weierophinney
You raise some excellent points. However, one thing to note: the W3C X/HTML validators do not accurately validate the specifications. As an example, the specifications actually allow for arbitrary attributes for any given element; the validators do not, and will flag these as errors. If we cannot depend on the tools, how can we possibly expect the browsers to follow suit?
Hi Samuel
Remember that the beautiful code is nice to have but not must to have. Sometimes it’s matter of something else than developers’ taste or previous indoctrination.
What would Google gain if their (lets take it as example) main page markup would be valid? I tell You what – few bytes more of code. Nothing more, as I am sure, this page is properly viewable by all mayor browsers, mobile browsers, browsers for handicap and much more. Now, multiply that few bytes by all page views per day/month/year – what you’ve got then?
HUGE bandwidth usage (I mean larger than it is now), which in consequence requires more servers to guarantee same quality of service, which needs more place to occupy, which demands for more energy etc etc. It is all massively, massively expensive.
Remember that: being stuck to standards doesn’t necessary pay off. Certainly not in that scale.
Twitter: @revellresearch
Samuel
I largely agree with your interesting post.
I really don’t understand why web designers continue to fail to produce compliant code and why many try to defend this!
Three thoughts (in no particular order!):-
(1) Pages with compliant-code often receive much better results in search engines, so why pay for web designers who damage your chances of being found…
(2) In many jurisdictions around the World, it is now a legal requirement to provide equal access to people with disabilities (who might be using specialised browsers). They have a much better chance of accessing the page if it sticks to the standards! Why would you want to operate or design a web site that effectively discriminated against disabled people?
(3) Imagine the public outcry if it was found that bridge builders were ignoring the standards in their profession! The same applies in many other professions – Why should web design be any different?
My article “Web Compliance? It is often simply unbelievable!” is along a similar vein to yours, although I concentrate on those that say they comply and then clearly don’t. I’m forgiving of odd errors, to err is human, but many such sites simply have thousands of errors!
Alastair Revell
Managing Consultant
Revell Research Systems
Twitter: @Samudary
Hello Sam,
This is a very good post. I do agree that we should spend more time adhering to standards in our code and web design, however, sometimes its not 100% possible. When I take a look at my webalizer statistics for some of my clients’ sites I notice that a good percentage of visitors are still using the troublesome IE. But I don’t want my clients to miss out on potential page views so I just design something that makes all browsers happy. I do think that its cause for concern though that a site like amazon could have over 1,000 errors, that’s ridiculous. They need to set a better example.
Anyway, keep writing, we need more bloggers in Jamaica on these subjects.