A client’s guide to HTML validation
HTML validation is one of those few technical topics that clients should really know just a little about, if only to keep their Web developers on their toes.
- by Peter Labrow
Despite my best efforts to describe technical topics to clients in plain English, I’m painfully aware that I sometimes inadvertently drop words into an explanation that only a Web developer would really understand.
In truth, such topics are often not even of interest to clients. As Web developers, we may well be fascinated by what goes on under the bonnet of a website (or under the hood of an American one) but clients are only really interested in the end product.
Know the language of your suppliers
There are some topics, though, which are helpful for clients to understand, because it helps them to make better buying decisions and also to assess the work of their suppliers more easily. Validation is one of those.
Most people are aware that their Web pages are written using a mark-up language called HTML. However, clients don’t usually want to understand or write HTML – they just want to get on with what they do best, and let someone else sort out that ‘HTML code’.
Therefore, clients assess the quality of the finished Web pages on how they look – at least initially. They will also, over time, judge their website on other factors, such as how it performs in search engines or how many people buy something from it.
Making more than just a visual inspection
Assessing a Web page on how it looks alone is, unfortunately, flawed. Yes, the pages need to look good, but, under the bonnet, they need good quality code, too. But how can a client tell if the code is of good quality?
Any decent Web developer will tell you that there is an easy way: validation.
Validation is a kind of grammar checker for HTML. You use a validation tool to see if the HTML is properly written.
Good Web developers validate their HTML as they work, and only supply clients with pages that contain valid HTML.
True, the reports that a validation tool generates may as well be Klingon to most people. But clients don’t need to know the ins and outs of why a page does or doesn’t validate – just whether it validates or not. So, this can be an essential quick check to confirm the quality of the work provided.
Why is validation important?
There are lots of reasons why all Web pages should contain valid HTML. Many of these are close to the client’s heart.
- It is a basic check of a job well done and shows that the Web developer both knows his/her onions and cares about the quality of the finished result.
- Having valid HTML helps to ensure that your Web pages will display consistently in all browsers.
- Valid HTML can be a contributing factor to getting higher search engine results. (It’s worth noting that there that plenty of Web developers who have some doubts about this, but it is typically only those who don’t produce valid code. They cite examples of many Web pages which are poorly written but still get good search engine results. However, valid HTML is only one factor in getting good results – search engines have had to learn how to index poor code because the majority of Web pages are poor. But they can do better, with better quality HTML.)
- It shows that the developer is most likely writing the code by hand, rather than using a WYSIWYG page layout tool – such tools can create very bloated code, and slow-loading pages.
So, pages that have valid HTML can load faster, be indexed better by search engines and display more consistently across different browsers.
How do I check if a page is valid?
There’s a free on-line tool, which most good Web developers use, to check any page pretty much instantly – it’s the World Wide Web Consortium’s Markup Validation Service.
You don’t have to worry too much about the individual errors, you’re just looking to see if your pages get a nice bright green (pass) or a nasty red (fail).
Plenty of websites do fail, but that doesn’t mean yours have to. Even popular websites can fail (at the time of writing, Amazon’s UK home page has an astonishing 1556 errors). Yes, Amazon does do well in search engine results, but there are other factors in play that you are not able to replicate – our advice is always to gain every advantage you can, however small, as they all add up.
Labrow Marketing and validation
We use the World Wide Web Consortium’s tool to validate pages, but we also use a more advanced off-line tool, CSE HTML Validator, which is a cracking piece of software and overcomes many of the limitations of the World Wide Web Consortium’s free tool. It only runs on PCs, which is a shame for me since I use a Mac, but it’s the one piece of PC software I use on a regular basis.
We also guarantee to all clients that all of their pages will validate, unless a technology is used which itself creates code that isn’t valid – though such instances are few and far between. If it doesn’t validate, we fix it, or explain why it can’t be done. (To be fair, this does not apply to some older sites, and we take it on the nose that those were created before the issues of validation weren’t as well understood as they are now – but we all have older work in our portfolios that could be improvised with hindsight.)