This week has been an interesting one in regard to the codebase of mortalpowers.com. I have increasingly become interested in the site as a way of helping others with issues, with finding new and interesting information, and exposing my ideas and projects to the world. As a part of this I have increasingly been working on the design, function, and SEO of this site. There have been several components overhauled in an attempt to make things easier to use, and to allow them to be more easily indexable by search engines.
Page titles and Page Descriptions
Using the Google Webmaster tools, I have systematically been following all reports of duplicate page titles and duplicate page descriptions. I started with Page Titles, ensuring that each of the article and content pages included a title related to the article. I also went through some of the more generic navigation pages and added a title that would indicate where the user is. In theory this will not only help SEO, but also make the site easier to navigate and understand for users, especially those with different accessibility needs. Originally going through page titles, I though that perhaps the page descriptions were unimportant, but recently I finished the page titles and began writing content for the descriptions, as I realized these are used by search engines to sometimes show a summary of the page, when specific page content didn't match the user's search.
One of the difficulties I have run into with page descriptions is that I'm not sure what tense/person to write these in. I have tried to keep all of the page descriptions simple and declarative, and I'm sure there is a standard out there I'm supposed to follow, but that will have to come another day.
I had a lot of legacy URLs, for example the imaging system were all under the name "posse", which was the name of the group I originally had photos of. I have been refactoring and redesigning the site so that all of the urls make sense. This means that "posse" is now "photography", which definitely more accurately describes what that section of the site does. Another area I have improved is with the article system. It appears that a few characters have slipped in over the past year that weren't removed from the URL string version of the article, meaning that things like single quotes were included, when they have the potential to mess up spyders or browsers that don't process them well.
The final piece I have been working on is to add a few 404 errors to pages where the user is searching or browsing. The best example of this is the new tagging system. Before with the tagging system, if you entered a tag that didn't yet exist in the site, you would receive a blank page that didn't tell you anything about what happened. This type of user interface leaves the user wondering "what happened? Did I do something wrong?". In response to this, if you try to view a tag that doesn't exist yet, you will get a 404 error, as well as a friendly error message explaining what happened.