(I found this one this morning and this roundup from the other week is also very good.)

It’s quite a specific analysis of his site, but it can be used as a basic framework for most sites out there:

First off, page titles (as i call it, the
This is pretty key. Most search engines use this as the heading for the search result listing, the link that you click to go to the desired site after performing a search. This is one of the primary places a search engine will look for keywords. However, it shouldn’t be too long as it will get clipped/truncated and it should make some sort of sense. I know this may seem obvious, but there should only be one tag on the page and it should always be inside the section of the page.

META tags:
Right, to cut through all of the confusion, the only ones you really really really need are the description and the content-type ones. The description should be different for each page and should be no more than one intelligible paragraph about the contents of that page and if possible not just a paragraph that is already written on the page.

The content-type is a little more confusing, but suffice to say as long as it looks like this on every page of yoursite, you’re ok.

Of course, on other sites, this needs to be considered carefully. Web browsers use a number of methods for determining the correct content-type of the document and if they’re mismatched, you may end up with the wrong one and certain characters will come out with extra glyphs, especially if you don’t use ANSII code for special characters (e.g. &123;).

The rest of the META tags aren’t overly used and in the case of the keywords one, ignored altogether. Any META tags should appear inside the [in here] tags.

Valid HTML:
This is extremely important to search engines. Clean code means it’s easier for them to read your site and suggests that it will render well in the browser, which you’ll score brownie points for. Code that isn’t where it should be will confuse the search engine algorithms and they may even give up indexing your site completely until it’s sorted.

This is a difficult one to achieve as there’s a lot that goes into this. It comes down to having a good basic design and sticking to it. One thing I will say: make sure there is no code or content floating around in between the closing tag and the opening tag or after the closing tag (except for the closing tag).

Headings

through

:
Headings are also really important. If you think about the basics of print for a minute (this is where all this comes from anyway): When you open a book it has an index giving you a quick glance at all of the chapter headings. If you go to a chapter, you see its title in large, bold text at the top of the page. Then the content relevant to that subject is placed underneath and is generally organised by subheadings and paragraphs. This is so we can follow the train of thought without getting lost and easily pick up where we were if we do.

If we apply this principal to the web, it becomes very natural, but also meets some requirements of the search engines. So having a main heading on each page (the

tag, there should only be one of these per page) that re-iterates the title of the page and then structuring any text into paragraphs of single thoughts, just like you learned in English lessons, will go a long way to improving not only the ease of reading from a visitors point of view, but also the search engines.

Flash:
For the most part search engines can’t read Flash content. A search engine basically sees what you would see if you did a “View Source” in your web browser. They use the text they see to determine what the page is about, how relevant and up to date it is etc etc. If that text is in Flash it won’t see it. If there’s any major bulk of text in a Flash file that plays on your site, it needs to come out and sit on the page somehow.

I don’t think you’ve got this problem as most of the flash you’re using seems to be image galleries, which is fine for the most part. There are alternatives to Flash which could improve your site in this regard, but it’s not essential.

Links:
Firstly navigational links on your site should be clear and steady. By this I mean that as you move from one page to the next, they should stay in the same place. They can also serve as a visual cue as to what page the visitor is on, so links that disappear when you’re on that page can be a little confusing.

Visitors should be able to get to almost any page from any page. So rather than having to leave a trail of breadcrumbs, they can simply see where they were when they read that really interesting part/saw that really good photo.

Secondly, links from other websites. Getting other sites linking to your website is another key from a search engines point of view. But rather than getting hundreds or thousands of links from websites all over, it’s better to have even just a few that are more relevant to your field of expertise. And the more natural the link looks on the other persons/company’s website, the greater the chance that it will improve your ranking. E.g.

Click Here! is not quite as useful to Google as Birmingham photgrapher portfolio or something similar. Can you see why?

If you can encourage people to link to your site or write an article about you or something like that, chances are it will be more natural.

Others:
Some search engines use a simple datafile to help identify pages on your site. It’s called a sitemap XML file. This is a bit complicated and techie, but setting one of these up can complement a well-delivered website and make sure that you tick all the boxes from the search engine’s point of view.

Also content freshness is an area to consider. Although I have found that this doesn’t have to be too dramatic, some changes every now and then help to keep your site on the map so to speak.

The points here are pretty obvious if you’ve been doing SEO for a while. But they need to be monitored to make sure you continue to comply.

There has been a lot of hype in design and developer circles for a good few years surrounding all of this, especially for proponents of the so-called semantic web – the supposed natural evolution of the web.

However, Tim Berners-Lee (seen by many as the father of the web as we know it today) has already suggested that this semantic web will only make up part of what we will come to call Web 3.0. How much of it will be the semantic web is yet to be seen (if much at all given it’s progress!)… more to the point, how much of an impact these impending technologies will have on the semantic web is a little hard to judge.

It strikes me that whenever we reach this next phase in syntactical changes for the web as we know it – in terms of it being an officially approved and ratified recommendation by the powers that be (some time around 2012) – and the point in time when it can be considered as globally adopted – probably within 4-to-5 years following that recommendation, similar to that of XHTML1.0 – will be half an evolutionary cycle.

If it does bring us anywhere close to the intended semantic nature of the web, it will, at best, be only half way there. So I’m going to go out on a limb and predict that sometime in 2015 we will confidently say we have reached Web 2.5.

Of course this is assuming we’re still here in a fashion. And that this stuff moves on apace. With the current fairly good awareness of standards and best practice, I believe that a small nudge from some big players may impact things for the better. Say Google adjusts various search algorithms to favour HTML5 sites in search listings… we all know that clients will notice and designers and developers will have to pay heed!

So it really is best to start now. Most of the changes (particularly towards HTML5) aren’t major. In fact, as can be seen, they should simplify our lives an awful lot! The problem is that there’s a lot of web out there to change. You can’t just change tags and roll – the implications are far greater: you have to consider CSS, the impact on any server side scripting used, browser rendering and their differences (especially for the new controls)…

This all sounds a little painful. For those using any kind of Web App platform with a good templating structure, this should be fairly easy: set up a new HTML5 template. The only complication to consider then is script-generated mark-up. And that should be tackled by the vendor.

Of course this is somewhat subjective as browser support is sketchy at best, so it’s hard to test any development in this area. Looking forward, we should be seeing greater support of these technologies in coming months. For now it’s probably best to glean what you can from the WHATWG and if you’re a developer building a CMS or other Web App you can probably start writing some test cases and replacement libraries just to stay ahead of the curve.