• Register
• Login

SEO Theory - SEO Theory and Analysis Blog

Algorithm analysis, Web community relationship analysis, SEO practices and techniques, industry news, etc.

• Home
• About SEO Theory
• SEO Glossary

Managing the SEO transition to HTML 5

Posted by Michael Martinez on January 23, 2008 in Advanced SEO

The W3C has published an early draft of the proposed HTML 5 specification. Now is the time for search engine optimizers to start looking at how to manage the transition to HTML 5. The implementation of the standard may be 1-2 years away but the SEO community does not handle standards well at all.

Many people in the SEO community latch onto and promote very bad ideas about standards. You can still hear a few voices squeaking about W3C-compliance, although search engine representatives have repeatedly said they don’t give points for W3C-compliance.

You’ll also still see plenty of arguments for using CSS instead of HTML tables for page layouts (which is a non sequitur - HTML tables were originally introduced to assist with page layout issues, not simply to organize tabular data, but CSS was developed to both enhance and extend the page layout options HTML provides).

These stupid, wheel-spinning arguments are time and resource wasters. You accomplish nothing to by “converting” anyone to your point of view. You certainly don’t improve your search engine optimization skills by advocating any particular “standard” or position.

However, search optimization has to share the load with accessibility and usability, both of which are directly impacted by the standards that Web site developers follow (or by the lack of adherence to such standards). Generally speaking, if it’s good for either accessibility or usability it will at least be harmless for SEO but oftentimes is very helpful.

I would take a Web designer who knows nothing about SEO but complies with accessibility to the letter over an SEO who has taken any presently offered SEO course without question, if I had to hire someone new for my team today. The search engine optimization industry is so confused and cluttered with myths, nonsense, and misguided good intentions today that the average SEO brings as much baggage to the table as knowledge and experience.

And this is where we have to look down the road, because the SEO community generally doesn’t recognize technical problems fast enough to offer credible solutions. For example, the new HTML 5 standard proposes that the old Hx header element hierarchy be replaced with a universal <h> element. A new <section> element will be used to create the stepped, layered, or nested hierarchy we have traditionally been creating with H1…H6 elements.

So instead of doing this:
<H1>Some Page Heading</H1>
  <H2>Sub-Section 1 Heading</H2>
    <H3>Sub-Section 1 Part 1 Heading</H3>
    <H3>Sub-Section 1 Part 2 Heading</H3>
  <H2>Sub-Section 2 Heading</H2>
    <H3>Sub-Section 2 Part 1 Heading</H3>
    <H3>Sub-Section 2 Part 2 Heading</H3>

We’re now going to have the option to do this:
<H>Some Page Heading</H>
  <SECTION>
  <H>Sub-Section 1 Heading</H>
    <SECTION>
    <H>Sub-Section 1 Part 1 Heading</H>
    </SECTION>
    <SECTION>
    <H>Sub-Section 1 Part 2 Heading</H>
    </SECTION>
  </SECTION>
  <SECTION>
  <H>Sub-Section 2 Heading</H>
    <SECTION>
    <H>Sub-Section 2 Part 1 Heading</H>
    </SECTION>
    <SECTION>
    <H>Sub-Section 2 Part 2 Heading</H>
    </SECTION>
  <SECTION>

That’s pretty ugly but it’s the wave of the future. Fortunately, at least for the foreseeable future, it will just be an option, not a requirement. However, the W3C standards community has a lot of support among Web Design tool makers and that is where problems will start to arise for the Webmastering community.

The HTML 5 standard will appear first in Web design tools that generate the code. Browsers will be updated to handle the code but many people won’t update their browsers for a long time. So many HTML 5 pages will not render universally. The people most likely to intentionally adopt the HTML 5 page coding standard will do so (in many cases) with a chip on their shoulder. We’ve seen this happen twice before.

The first wave of “Web page design with attitude problems” occurred when Microsoft launched Internet Explorer as a competitive browser. Mosaic and Netscape dominated the Web client market, although Lynx (a text-based UNIX browser) was still pretty popular in educatioal institutions and large UNIX shops. Microsoft extended HTML features and added support for those extensions in Internet Explorer (Netscape had done the same thing).

It became common in the mid- to late 1990s to find Web pages promoting Microsoft Explorer with little tags that said, “This page looks best in Internet Explorer. You can download it here.” Some Web site operators all but broke the Netscape browsing experience by fully implementing Microsoft’s features.

As the browser wars cooled down and Internet Explorer became the dominant browser those old “This page looks best” markers became mostly a thing of the past. But then people — unhappy with the Microsoft experience — proposed developing a new browser technology. Enter Mozilla, which though Microsoft supports Mozilla is nonetheless seen as an option for developing alternatives to Microsoft. And that brings us to Firefox.

A lot of people in the SEO industry prefer to use Firefox over Internet Explorer. The readership of this blog has — until recently — favored Firefox over Microsoft’s browser by a margin of almost 4-to-1, the exact opposite of the broader Web surfing community, where Internet Explorer still dominates browser share with around 75% of the user base.

The extreme prejudice in favor of Firefox has brought many SEOs to the brink of self-destruction — that is, when I visit their sites with the latest version of Internet Explorer, their sites hang. Firefox handles their Firefox-friendly designs just fine. But the majority of Web surfers see some very ugly Web sites from the SEO community and this is one of those rare occasions where ugly don’t work.

So what does that have to do with HTML 5? Early adopters will often take a “Support it or leave it!” attitude with their Web sites. If you’re in the business of building traffic and increasing conversions, there is no room on your shoulder for that kind of chip. If you want to develop only HTML 5 sites for HTML 5 browsers and no one else, you’re free to do so but you’d better hope that the HTML 5 loving community builds some HTML 5 directories and search engines, because Google, Yahoo!, Microsoft, and Ask won’t favor HTML 5 pages.

Now, a more direct issue for search optimizers is that as we or our clients adopt new Web design tools, those tools will sneak HTML 5 code into our page designs. The Hx headers may start vanishing as H/SECTION combinations are inserted for unwitting page designers. If that happens, we have to hope that the search engines will have taken the H/SECTION standard into consideration. Otherwise, people who have depended too much on their H1 headers for on-page optimization will appear in SEO forums claiming that they lost their rankings but “I didn’t change anything.”

The more seemless the Web design transition proves to be, the more likely people will not notice that their old Hx headers are updated to the H/SECTION headers.

This will only be an issue if the search engines are caught flat-footed.

Historically, search engines attempt to look at the emphasis placed on words inside the document. There is a little — though not much — publicly available technical literature that explains some ways in which this may be done. These documents are fairly old and unless a search engineer discusses these points publicly we have no way of knowing what is really happening. But when you’re indexing documents that come in a variety of markups (and which are not marked up at all), you have to make broad allowances for emphasis.

So the good news about HTML 5 is that the search engines have a lot of experience at indexing documents created according to many different formats. They still index HTML 2, HTML 4, and plain text documents just fine. They’re indexing .PDF files, .DOC files, and even Flash files. We can reasonably expect the search engines to adapt to HTML 5 as they see it come online. But how many HTML 5 pages will be indexed improperly before the search engines have tooled up for the new standard? We don’t know.

Another issue that the HTML 5 standard presents is how we manage our page heading structures. If done right you see the hierarchy as it should appear. But some people intentionally break the structure and some people unintentionally break the structure. Neither group is going to be happy if their pages are imported into an HTML 5 design tool that converts everything for them without asking or explaining what it’s doing.

Emphasis is an important tool in your on-page factoring. There are more ways to emphasize words in a document than most SEOs believe or care about. Anything you do to make a word stand out above the rest helps convey the impression that the word is important to the document. Good search optimization relies upon more than one form of emphasis. So does good page design and presentation.

If a transitional issue arises during the early adaptive phase of HTML 5’s introduction to search optimization, it may take some time for people to realize what is happening. HTML 5 will be so new to most Web designers and optimizers that people won’t think to look under the hood to see what is going on.

My advice for search optimizers is to take a wait-and-see approach. Primary Web content should remain in HTML 4 format for at least a year after HTML 5 documents begin appearing. When HTML 5 becomes usable, SEOs need to create HTML 5 content and see how well it performs in moderately competitive search results without relying upon HTML 4 alternatives for emphasis. The testing will be sporadic and the quality uneven but the more testing that goes on, the better the SEO community will become at managing the transition.

HTML 2 has not gone away, but HTML 4 has become the de facto page coding standard for the business Web. HTML 5 should eventually supersede HTML 4 just as HTML 4 displaced HTML 2. We need to look at each new proposed method of asserting emphasis. We need to test these new elements and structures in safe documents that are not flagship documents.

But most importantly, we need to keep our opinions on the back burner. Another round of “X versus Y” arguments won’t be any more productive than the last round. A lot of Web sites have been unnecessarily redesigned because SEOs felt that W3C-compliance was necessary to improve rankings. While accessibility and usability may have improved for some of those sites, a lot of once well-designed sites are now virtually useless to the non-Firefox using Web surfing community.

The SEO industry not only needs to correct that mistake, it needs to make sure it doesn’t make the same mistake again. And if in the transition to HTML 5 some people in the SEO community launch into “X is better than Y” diatribes, we’ll all know who NOT to listen to.