Posted by: jimmydlg | January 25, 2009

Meaningful Markup

I’ve been familiar with the concept of creating pages with meaningful markup in mind, even though I in recent projects I haven’t put those concepts into practice much. It’s not that I disagree with the concept, it’s just never been a point of concentration in the work that I’ve been tasked to complete.

I do, however, forget (as Denis Defreyne points out in his article about meaningful markup,) that not everyone (and everything) can make use of all my carefully crafted styled and designed sites. Search engines and users that don’t have a graphical or even full desktop browser could possibly benefit from having a site that concentrates on seperating content from delivery by avoiding certain legacy tags and keeping the content of the page in mind.

Robert Nyman suggests removing all the CSS, scripting and possibly even images when examining the content of your page, to see if the content is well organized and connected, and even though I had never considered this before, I could see how this could be useful. My typical approach to reviewing the completion of a website is to take a step back from it, and see how the page flows graphically. This technique could certainly help me evaluate the content as well, so that it is both presentaionally pleasing, and semantically appropriate as well.

While one of the tenants of meaningful and semantic markup is to identify content with meaningful class names (when using CSS,) I don’t completely agree that maintaining some basic presentation references isn’t warranted in some cases. For example, I commonly define normal text as the class “normal”, and this is probably a habit I won’t be breaking for a while. Designers can get carried away of course with naming classes after formatting, but to me, saying that text should appear “normal” is both presentational and semantic. While I may discover a better way to identify normal content on a page instead of defining it as “normal,” what I can’t see doing is blanketly assigning default font style and size definitions to as many tags on a page as I can think of. I actually enjoy having to be explicit about how I want certain content to appear, but who knows, maybe that will change in time.



  1. Your comments about “normal” are of course your option. I would suggest however that normal is not necessarily vital in that body copy can be defined without a name, simply by using the body tag to define the consistent type data for “body copy”. While your normal may have some other meaning, it may be a class that needs to be written in HTML (class=”normal”) whereas the body selector can accomplish the same thing with no class or extra markup.

  2. Well it’s interesting, I said that, but as I am designing my newest layout, I have yet to use Normal. So perhaps it wasn’t as necessary as I thought 🙂

  3. I tend to use the “body” tag to assign universal attributes, such as what I think you’re describing in using the “normal” class. I find it much easier than having to constantly assign classes, especially when dealing with something I would consider my “default” font.

    It’s probably my previous classes with Wayne that have trained me this way, but I do find it helpful to think of how to the content breaks down by solely looking at the text first, before I add CSS, scripting, or images. That way, all the style or scripting elements that follow from that point are originally based on the content hierarchy I made with the raw content.

  4. I too take Ian’s approach and give my body tag all the common attributes to in regards to typography, probably as a result from previous classes.

    I haven’t quite yet tried the approach of completely forgetting imgs and other imagery on the page, there something always kinda tugging at me to make sure it looks good or as good as I can get it. but I suppose I can see where that can help with the communication of your content.

  5. This is an interesting discussion that I hope we all continue. I am not trying to impose a “style” of coding in classes, but I realize that can happen without knowing it. I think there is a way of presenting material, especially XHTML/CSS, that helps people learn the core fundamentals, but I never try to say this is the only way! I believe in semantic/well-formed code because of the developer community who have taught me the advantages of working together to improve the Web as a global communications process, and I have been confident in what I understand about this movement and the professionalism involved, but I don’t proclaim to push it on anyone. Get the core skills down, understand the environment we are working in, and contribute your best to it!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: