jamin.org

For the first time since the Internet was invented, Carnegie Mellon redesigned their website. Hurray for breaking out of 1996!

I applaud the grand undertaking of overhauling a university website. Having worked at the University of Pittsburgh as a web developer for three years, I understand that this is no small or simple matter.

There are numerous obstacles to overcome, politics to play, and too many people to please. Given my knowledge of those difficulties, I commend the result.

At first glance, the large photography helps the site look interesting. The main area is all Flash, and is a decent use of the technology to pull of subtle effects.

However…

Code Like a Platypus

Being the web standards nut the I am, after a minute I dug into the source code. Was I surprised to find an inflexible mishmash of CSS and tables? No. Was I disappointed that my cutting-edge tech school hired someone (Ripple Effects) that doesn’t know how to build without tables? You betcha.

I’m guessing the higher ups at Ripple don’t know about web standards, so the developers are not being forced to any standard, as long as the site appears to work; but the developers have heard enough about this CSS thing to throw in some div tags here and there, but not enough to actually achieve the layout they desired.

The resulting code is a bunch of gobbledygook.

These developers are not clueless, mind you. They’ve got some nifty Flash with XML and CSS going on, they’ve employed SWFObject to display alternate content if visitors don’t have Flash (though their alternate content has some problems), and they’re messing around with DOM scripting.

With all that going on, the lack of full CSS layout is a mystery, but unfortunately not unusual.

Just for fun, try turning off Javascript and images. I know, who does that? Me!

Light as an Elephant

It’s typically a good practice to keep your site light and speedy, like a race car. This site has six overweight passengers crammed inside, and a few sandbags in the trunk for good measure.

While there is an ever increasing trend toward DSL and cable Internet service, people do still use dialup. This site doesn’t care about those people.

• Total recommended size for 56 Kbps modem: 30 kb (8 seconds to load)
• Total size of home page files: 365 kb (73 seconds to load dialup, estimate)

The page loads in about 10 seconds for me using cable modem. Not bad, but these numbers are:

• 48 images: 126 kb!
• 5 scripts: 162 kb!

As a comparison, the University of Pittsburgh home page, which I built, is 58 kb total: 30 images at 24 kb; 3 scripts at 9 kb.

I’m not holding up Pitt as the standard of excellence.

Looks Like a Swan

Yep, it looks better, and it appears to work. But because of the above negligence, the developers failed.

General Take on the IA

I have some questions about the information architecture. But as I am not privy to the user research and strategy, I cannot fully comment. Though the Post-Gazette article about the launch indicates the guiding principle was people and stories.

I will say that I like that information is chunked in an attempt to make navigating more manageable for visitors. But does it? There’s a look going on, which may be overwhelming.

It’s interesting that Search is the first thing beneath the Carnegie Mellon logo. Does this mean there’s an expectation that visitors will not know how to navigate the site and will rely on searching to find the content they seek?

I feel a bit sorry for News, tucked away as it is beneath all the navigation, making room for the themes. I wonder if people will really submit theme ideas.

I also wonder about the font sizes, especially on the subpages. With all the large images and large main navigation links, the body text does not appear very readable (but who reads?). And check out the line length on the Research page.

What Many Are Calling Web 2.0

“In what many are calling Web 2.0, companies, nonprofits and universities are spending billions of dollars redesigning their Web sites, with varying degrees of success.”

That’s from the Post-Gazette article. Read it again. Did you catch the revelation? Web 2.0 is about spending billions to redesign websites, and possibly utterly failing.

Ending On a Good Note

I’m harsh, I know. Overall, I’m very happy to see the site has finally been updated. It’s a huge improvement, and a step in the right direction.

I just hope it doesn’t take another 10 years before the site is updated.

Shaun Inman, a major force in sIFR, the man behind Mint, and a very nice guy whom I had the pleasure of talking to earlier this year, once again redesigned his site. And it’s damn interesting change.

While his process of designing in the browser first and then going to Photoshop for only minimal pixel art ain’t much different than the very site you are reading (I’m so ahead of my time), his algorithm for color coding content according to its relationship to the present is quite original.

I found it intriguing that he is thinking about the proliferation and inevitable obsolescence of content on the Internet and the challenge of communicating relevancy of information to visitors that may not know the whole history of development who happen upon older techniques or opinions.

I’m not sure this is the best answer to that question, but it is a fascinating solution, worthy of consideration.

What I didn’t say yesterday was that we had to choose an industrial design firm for our one-page write-up. I looked at a half dozen sites before deciding. The determining factors for the company I chose were the site didn’t totally suck, and they had photos of the products and text I could grab for quotes.

Unbelievably, one of the sites I looked at only had a front page that consisted of one image and no content, and another had multiple pages but was still built as all imagines. (Turn off the images, and poof.) It amazes me that these arguably top-notch design firms care so little—or perhaps understand so little—the impact their web sites have on their reputation, and how ineffective their sites actually are.

I’m not saying industrial designer should understand web design. But this is an example of a larger observation. The more sites I look at and clients I work with the more I see a ubiquitous lack of understanding of how and why you should create an effective web presence. And this means there is either a huge opportunity for web design firms or gigantic void (can you have a small void?) in the current ability of web design firms to help companies build effective web sites.

Making Candle Holders

Our objective in choosing a industrial design firm is to create a tea light candle holder from paper that can be mailed in a flat 9″x12″ envelope and represents the spirit of the firm.

Yes, it does sound like a fire hazard. But we created tons of prototypes today and put tea lights in all of them, and none caught on fire. Fun project. Of course, when isn’t playing with fire in a classroom fun?

Today concluded our week with Karen Moyer and typographic hierarchy. I’m not sure if I mentioned this before, but we will have a new instructor who will explore a different aspect of design each week. Next week has to do with 3D objects, I believe.

After class we were treated to lunch with Dan Boyarski, head of the School of Design. During this lunch, while Dan was chatting with others, one of my peers asked me about my thoughts on taking a web design class to learn to build web sites. She wasn’t sure if she needed to learn HTML or what. And I started telling her about my distrust of web design courses, for these reasons:

• Web design (and by design I also mean development) moves very quickly, so instructors need to be actively engaged in web design, which I don’t think is the case.
• If you can be assured of finding information about anything on the web, you will undoubtedly find tutorials, articles, and examples about web design, so you don’t need a course.
• Most non-University web design workshops (Event Apart and the like excluded) are business driven and more concerned about getting your dollars than what you take away.

But before I completely dissuaded her from taking a web design course, I turned to Dan and asked him what the courses were like at CMU. What he said encouraged me. And I appreciated that he seemed to agree that designers need to understand the code behind the design, even if they don’t do the coding itself.

Then Dan asked me how I would teach a web design course. I said I would take a holistic approach that started with a focus on the strategy and goals of a site along with an understanding of the audience, and show how that drives everything else: information architecture, design, technology.

Only later did I think about how I could have given an answer that was completely focused on markup and standards, leaving the sum of web design for a broader class. Regardless, Dan seemed to like my answer. And it got me thinking that I could very well teach a broader web design class.

I’m not sure if that would be a possibility or if that’s what Dan was fishing for, but I find the idea very appealing. Presented with the opportunity, I would accept.