Why do so many web development companies websites fail W3C validation?
So many I have looked at have masses of validation errors on their pages, often over 100! Few even seem to bother with a doc type! Why?
| SitePoint Sponsor |
Why do so many web development companies websites fail W3C validation?
Why do so many web development companies websites fail W3C validation?
So many I have looked at have masses of validation errors on their pages, often over 100! Few even seem to bother with a doc type! Why?
Because browser vendors bend over backwards to accommodate all sorts of poor-quality markup. Thus these sites 'work' even though the code is invalid. Of course it may not work properly for the minority that doesn't use IE or Firefox, doesn't have 20/20 vision and no mobiility impairments, etc., but ignorant designers/developers happily ignore them.
Also, most developers/designers have never read the W3C specifications. They've learned by inspecting the source code of other (poorly written) sites, or by simple trial-and-error.
Birnam wood is come to Dunsinane
No idea, but I wouldn't want to be hiring any of those companies.
Though one (admittedly poor) reason I have heard is that your own website doesn't make you the money, it's the websites you make for others that make you the money.... you can't go charging clients for work done on your own site, therefore it gets sloppy.
I think that's a load of crock, but that's one possible reason
w3c specification is really not that hard,for xhtml you only need to put some close tags.
and don't use tag that doesn't supported by w3c standard
I had a similar conversation with a client the other day.
The web design industry is still young, you will struggle to find a web designer that's been at it for over 10 years, as a result the industry is still suffering from growing pains.
Your Average Joe web designer (I know I'm making a huge generalisation here) is a one man band who's self taught by using online tutorials and viewing source code, as a result he uses tables, and probably doesn't know what a doctype is, let alone what it's there for.
Slowly these designers tend to fall in to one of two categories.
1. I'm making money now, why bother learning web standards.
2. This web standards stuff makes real sense.
At the moment group 1 is the majority, but things are changing.
Technology is dominated by two types of people:
those who understand what they do not manage,
and those who manage what they do not understand.
It does make it easy to tell the amateurs from the professionals - for those who know how to validate a page.
Stephen J Chapman
javascriptexample.net, Book Reviews, follow me on Twitter
HTML Help, CSS Help, JavaScript Help, PHP/mySQL Help, blog
<input name="html5" type="text" required pattern="^$">
I suppose that depends whether you class professionals by skill or success. I was looking at a web development company who had invalid pages but have been trading successfully for 12 years!Originally Posted by felgall
AND, they charge a LOT as well. It’s a shame people don’t learn a little more about web design basics before they get out their wallet.
By attitude.
Birnam wood is come to Dunsinane
Of course they charge a lot - they would need to spend a lot longer working on maintaining existing invalidly coded pages than they would have to charge if they had the pages written properly in the first place and therefore need to charge a lot more because the same work will take them a lot longer to do.
Stephen J Chapman
javascriptexample.net, Book Reviews, follow me on Twitter
HTML Help, CSS Help, JavaScript Help, PHP/mySQL Help, blog
<input name="html5" type="text" required pattern="^$">
I guess that depends on the definition of "professional" - as "someone who is very good at web dev", or "someone who makes money out of web dev".
Many web design companies just use tools like Dreamweaver, and you get what it produces. If it doesn't validate, or doesn't have alt tags for images, or doesn't use external CSS files, that's just the way it goes, and the web developer can't do anything about it because all he knows is how to make the page look nice in Dreamweaver or whatever. That's not necessarily the fault of the tool, maybe just the devs lack of knowledge of how to use it, or the company using an old version, or whatever.
(Note: I'm not having a go at Dreamweaver specifically, it probably applies to other tools as well)
http://www.firenza.net - my homage to a car from the 1970s
My understanding is that Dreamweaver can be configured to produce pages that are standards compliant. Those using it without adjusting the settings appropriately have demonstrated a lack of knowledge of the basic tools of web page creation.
There are lots of so called professionals in most industries who make a lot of money without being competent in doing what they do. They don't tend to last long in those industries where their incompetence is most obvious but can easily last for several years in industries where what they produce appears to work. Eventually they get caught out.
For example, imagine someone builds you a 10,000 page web site without following the proper standards - obsolete tags, tables for layout etc. Now you decide you want to make a minor change to the layout of the page by moving something from one spot on all the pages to another. With properly written pages that might require two or three lines to be changed in the CSS but the person who built the site has to make major changes to all 10,000 pages instead, A five minute job for the competent web designer is a five month job for the incompetent one who will obviously need to charge accordingly since they will not be able to survive the five months on the couple of hundred dollars that the competent designer would charge for the five minutes they would need to make the same change to their standards compliant work.
Perhaps I should have said that it distinguishes the competent professional from the amateurs and incompetents.
Stephen J Chapman
javascriptexample.net, Book Reviews, follow me on Twitter
HTML Help, CSS Help, JavaScript Help, PHP/mySQL Help, blog
<input name="html5" type="text" required pattern="^$">
Yes, with a lot of care.
Standards compliance is one thing – layout tables validate just fine – but lean, semantic markup is something else. It's far hard to accomplish with Dreamweaver, if you work in the design view.
But then it's probably more work than handcoding.
Birnam wood is come to Dunsinane
because most of them don't think it is necessary to validate to W3C. Even I don't find it necessary to validate to W3C. Most big websites won't validate to W3C if you try them. But ya if you wont to pay extra for valid codes as well then you can go for companies that valid to W3c. I personally wouldn't mind if it has invalid HTML or CSS as long as it displays the same in all browsers.
Why should you have to pay extra for having the job done correctly? Do you think it's fair to charge extra for copywriting if you have to deliver correctly spelled text?
So you would'nt mind hiring a copywriter that made lots of grammar errors and spelling mistakes, as long as your readers could decipher it?
And why does a site have to display the same in all browsers? Do your visitors often use lots of browsers in parallel, looking for discrepancies? And wouldn't this make for rather boring sites if they have to look the same in Opera/Firefox/Safari/IE as they do in Lynx?
Birnam wood is come to Dunsinane
Two words, legacy and evolution.
Huge sites that have been around for a long time are usually plagued with code and functionality that has been lying around for perhaps five years or more, which poisons the end product and hampers the ability to evolve with web standards due to the fact that even attempting to refactor all of that legacy code in a way that lets you meet standards is ridiculously complex and could cost insane amounts.
I just checked a page on the company I currently work for and it has 392 Errors, 2 warning(s). Yes, it's that bad, but I'm sure there are many organisations with similar difficulties. It's really not as simple as "ok guys, let's make our pages validate".
Bob, I think it's quite understandable that sites that have been around for a long time have poor legacy code. However, there are still many brand new sites that don't validate. I find it more difficult to excuse those.
Birnam wood is come to Dunsinane
I was about to say the same thing.Bob, I think it's quite understandable that sites that have been around for a long time have poor legacy code. However, there are still many brand new sites that don't validate. I find it more difficult to excuse those.
Yeh, totally agree with that.
B'coz they sucks
Best Joomla Web Design and Development Agency :-
PSD to Joomla :-
I agree with most of the sentiments above: there's no reason in most cases why you can't have valid, semantic code most of the time. It's not hard and I would expect it from anyone with a modicum of professional pride.
However, one of the reasons the W3C dropped the requirement for validation from WCAG 2.0 is that they presented the argument that sometimes invalid code can provide a more accessible experience - for example embed is well supported but not valid.
So in most cases, I would agree that invalid code is not a good start: but invalid code does not necessarily equal doing a job incorrectly.
Jack Pickard | The Pickards
I don't see how bad copy writing can be related to not validating with W3C standards. If you are a good coder who knows your job and is up to date, most likely you will have a coding that validates anyway. Say a website uses table format instead of div. So what as long as it has useful content. Not everyone is going to see your source code & say oh it doesn't validate to this standard or that & care if it is HTML transitional or strict.
As for the same display across browsers, i mean that i don't want my page to display a big heading or horizontal scrollbar in one browser and ok in other & stuffs like that. Again this has only to do with poor coding, not validating to W3C standard doesn't necessarily mean that your site is incorrectly coded.
Validation is essentially spell-checking for your code. It should be as natural for a professional coder as spell-checking is for a professional writer.
Tables are as syntactically valid as divisions. They may be semantically incorrect when used for layout purposes, but that's not something an HTML validator will detect.
The validator will, however, detect your missing </table> tag which may or may not cause incorrect rendering in various browsers.
Birnam wood is come to Dunsinane
1. Well a lot of the sites on the web are old and outdated
2. Poorly Coded
3. Hacked together
4. And all of the points everyone on this thread have made as well
I think every web development company out there should pick up a copy of "bulletproof web design" and make their designers read it. i see it too with the web design companies in my area. tables for layout, style attributes on the body tag, empty alt attributes on images and sometimes none...etc. etc. and a lot of dreamweaver generated code and imported fireworks nav bars and what have you.
I really don't think there's any excuse for the bad code and implementation out there. I didn't go to graphic design school or anything so my sites looking good are a work in progress, but I code the right way: appropriate semantic markup, separation of content and presentation, accessibility always in mind. And all I did was read some books.
Yeah that is a really good book. Also, check out Tommy and Paul's new book 'The Ultimate CSS Reference'.
Posting Permissions
Bookmarks