Noah and Parker’s graduate alma mater, Tufts University, topped the “gold medal winners” list, and while the description leaned heavily on Twitter and Facebook for all contestants, both EduDemic and collegesurfing.com specifically cited Tufts’ “Spark” project. Parker was part of the original Spark team as part of his graduate work, hacking blog software and working on the XSLT bridges Spark used to tie together the project components.

Design by our friend Justin Zucco, and I’ll post a link to the live site (Balmat Law) when it goes live in a week or two.

http://www.youtube.com/watch?v=mSC84CUx_Uk

By way of example, let’s consider what happens when you search for “web development” near Amherst, MA. Google Maps shows results ordered by proximity to the center of town, and guess which web development company has an office closest to the center of Amherst?

When the Amherst College development offices move on-campus, there will be some closer-to-the-center addresses available, just in case you thought geography was irrelevant in today’s virtual world.

We’ve mentioned our work on FlyFi before; the designers, Corey McPherson Nash, recently sent out a press release highlighting their role in the project.

Another project we worked on over the winter recently went live when Running USA launched their new site. Running USA, a trade organization promoting the sport and industry of running, started down the site overhaul road nearly a year ago. We consulted with them over the summer, assisting them in identifying the major issues to be addressed by a new site, drafting the eventual RFP, and advocating implementation of particular technologies. As the project moved forward, we worked with the project managers (a company apparently so busy they don’t have a completed website yet) to turn their wireframes and comps into valid templates.

In both cases, our deliverables don’t look like much in a desktop folder: a few .html files and even fewer .css files. But being able to count on the templates to deliver (and across browsers – thanks, Litmus) isn’t measured in file counts.

PHEA, as we call it in the office (any title that long needs a TLA, even if it’s a FLA) is run by a former colleague of mine from Runner’s World, Alisa Bowman, and it’s WordPress at the core. Beyond that core, however, PHEA uses several plugins, a custom page configuration, and a very custom theme. Wrangling all those players became our job, with the primary goal of shifting the theme from a three-column layout which left the content column too narrow for embedded video to a two-column layout. The result is easier to read and places the reading focus where it belongs: on Alisa’s columns, not on the sidebar material.

We’d love to claim credit for this, but actually Alisa is that rare and delightful client who knew exactly what she wanted; she just didn’t know how to make it happen. We demonstrated potential changes on a staging installation on one of our servers, and when they were approved, used Subversion to push theme changes out to the live site.

We have a lot of respect for WordPress as a lightweight publishing system which is easy for non-technical users to pick up, but PHEA demonstrates that for users with a clear idea of their site’s function and the patience to try a few different options, it can also be a powerful tool for creating a rich site which is worlds apart from a cookie-cutter blog.

The first use of anchors is rare, nowadays. Every page has a default anchor, named “#“, which is set at the top of the page, and authors rarely set additional ones. They’re most frequently seen on Wikipedia pages, of all places, where subsections of the larger page are all anchored.

The problem comes in styling. What happens is that the text which is wrapped as an anchor gets styled as though it’s a hyperlink, even though without an href= attribute, it won’t work as one. This discourages a lot of webmasters from using anchors. One typical stunt is to not wrap anything at all, to create the anchor as <a name="anchorname" /> or <a name="anchorname></a> (the latter style is how Wikipedia does it). This isn’t available, however, to people using some kind of WYSIWYG editor in a CMS.

The real solution is to properly style a tags, and by “properly” I mean “not the way the default browser styles do it.” That means clearing styles from the a tag, and applying link appearance only to the :link, :hover, :active and :visited pseudo-elements.

Here’s how we did this for a recent project:

• Style default a tags to look just like all other text:

a {
  text-decoration: inherit;
  font-style: inherit;
  color: inherit;
}

• Now define the a:link pseudo-element to look the way you want your hyperlinks to appear. This pseudo-element is often ignored, but it’s the key to not applying link styles to your anchors. (We apply the same style to the :visited pseudo-element.)

a:link, a:visited {
 border:none;
 text-decoration:underline;
 font-style:normal;
 color:black;
}

• Style the other pseudo-elements, because otherwise they won’t show differently from usual text (i.e. if :hover isn’t styled, links will appear to “disappear” when users mouse over them).

a:hover {
 text-decoration: none;
 color: #a3211f;
}

There’s no real trick here: all that’s involved is respecting the full definition of the a tag, and all the states allowed by its pseudo-elements.

We can’t take credit for any individual aspects of the HitFix site; what we did was work as a team with HitFix’s internal Rails developer, Elliott Blatt, to make the site Beta-ready. This meant taking Photoshop demos and making fully-functional pages from them, working at all levels of the Rails stack. (We’ve been joking that our work on this site was in defiance of Brooks’s Law.)

Elliott has been a thrill to work with, having more experience with Rails than either of us, and we were surprised to find him learning from us almost as much as we were learning from him. What we’ve learned from this site is significant as well; it’s impossible to detail all the small conventions we’ve learned from Elliott, but one big example was learning HAML and its CSS sister, SASS.

It has also been an experiment in decentralized development, as some of the HitFix team is in California, Elliott is in Cambridge, and we’re in Amherst.

We’re still in active development with HitFix, but as I mentioned before, the beta is up, so if you’d like to take a look, you may find it interesting.