When link lists attack!

* An open letter to Jakob Nielsen, from Design by Fire.

Let’s end vicious punditry and promote good design from day one of a product. You can’t bolt on usability.

* Syncato, a very cool, XML-based CMS that really shows what data-driven programming can do.

Kimbro Staken, its creator, wrote a good article about the design of his software for XML.com.

* Hypergene wanted RSS from Tidbits. They said their CMS didn’t grok RSS. Adrian gave them RSS… but no details on how he did it? Come on, Adrian we all want to know.

Edit 1:30 p.m.: I had to add this entry on The complexity principle, well written, well though out, and well worth reading.

Posted in Web design | 2 Comments

Google goes local – forgets sort?

But a comment… what’s up with Google’s “new” local search (in case you are Andy Rooney).

It’s a great idea, but what interface designer fell asleep at the wheel… on the results page you can’t sort by distance.

You can narrow the results to certain radii, but I can’t actually sort by the distance column… am I the only one who thinks this is a no brainer.

I did a search for chinese and my address trying to find a good place to walk to, but I can’t sort the results. I can view the map, I can mentally catalog the distances, but that’s not the same as sorting and then immediately seeing the closest restaurant.

And in closing, a request to Google, or anybody listening… would you please make your next project a mass-transit directions site, because the one I have to use stinks.

Posted in Web design | Comments Off on Google goes local – forgets sort?

A case against sidebars

I was reading this good article about software commoditization over at OSDir.com, today and I paused when I hit the article’s sidebar on the use of the word commodity in Shakespeare’s works.

I was torn… a ‘graph to the left me, a ‘graph to the right of me. The sidebar was longer than one screen (and what is one screenfull anyway?), so I decided to continue reading the piece, but I didn’t come back to the sidebar.

The short story, embedded in the text of a large piece, has been a staple of publication design for a while now, but I don’t think it makes the transition to the small screen very well.

With printed material, my eyes merely need to scroll down the column of type until the end or the jump, and then I can move back to the sidebar. When I’m done with that page I’ll flip the page and repeat the process with stories that catch my interest. (Oh, if the Web had such a simple interface…).

But on the Web (or any screen-based medium), I actually have to scroll up.

There’s been research that shows users prefer scrolling to paging, and I think between that behavior and the prevalence of scroll-wheel mice, there’s no doubt that you’re better to have your reader scroll than page.

I think there are two strikes against embedded sidebars (unless they’re very short — like 2 graphs short).

1. The first is interest.
If I’ve gotten to a story-level view of an article, then there’s probably a fairly high interest level on my part in the article… I drilled down through all your index pages, so give me my content.
Presenting related content is good, but in the format of a sidebar, I’m torn between the article I looked for and the new content.

2. The second is physiology.
When we’re presented with a single column of type, our brain and eyes are ready to start at the top, and go downward in a sideways motion (take your pick).
When I get to the bottom of a single-column story page, I’m ready to move on, it just feels unnatural to move back up to read.

Perhaps the best solution is to provide a list of related links, including sidebars, right at the top of the article.

In addition, sidebar-style content — content that really should be attached to its main bar — could be placed at the end of the story.

On a single-column, single-page story, that’d be the logical place to put it.

Posted in Web design | Comments Off on A case against sidebars

I hate client-side scripting!

Hate’s a strong word, but appropriate.

I was working on a quick little script to show/hide the answers to a quiz (not going live till Sunday) that a co-worker was creating.

She wanted to do this with a pop-up window, for each answer to tell if you were right or wrong. I told her to let me have a stab at it since pop-ups are evil, I had been looking for an excuse to do some Javascript/CSS funness, and, did I mention, pop-ups are evil.

So a quick little script to check an item’s display property, and it’s all done except for the testing right?

Well everything worked OK in every browser I could test, except IE6/PC. Mr. NN4 Jr. there wouldn’t behave.

Oh, it’d change the style all right, but it’d still send you on the href specified in the link.

I couldn’t figure out why. I Googled, I Evolted, I screamed!

I couldn’t find any circumstance under which IE6/PC would ignore a “return false” statement — which negates any href in an anchor element when returned by an onclick event.

Then, on a whim, my eyes darted to the tray.

I have seen the enemy, and his name is Free Surfer.

It’s a little pop-up blocker that works in IE/PC… and, so it seems, also forces IE to follow a link despite a “return false” from a Javascript function.

I disabled it, my script worked, all is well.

But how many other pop-up blockers, or browser “enhancements”, do this, too. If it’s alot then that’d block a very nice, very usable solution… to provide some DHTML for those who can use it but also provide a backup URL for those who can’t.

Edit 4:17 p.m.: Opera 7 obeys both the script call to change the CSS and the link.

Posted in Web design | Comments Off on I hate client-side scripting!

Un-fixing your layout, fixing your CSS

MediaSavvy brings up two points about fixed-width layouts:

1. Fixed-width layouts should be flush left: I’m not so sure about this one. When I’m going to a new site I don’t expect any of that sites UI elements to be in similar places as the last site I visited.

It can be a bit jarring, I guess, to see the content move to the center. That said, if you’ve got to be fixed width, I say center your design, the white space will be balanced which is more eye-pleasing and harmonious than the giant backwards-L you get with a flush-left design.

2. Eliminating fixed-width layouts is also better because they give designers the illusion of more control of presentation than they really have. Preach on! Say it with me now folks… the Web is not print, the Web is not print. They are different mediums, they are different mediums. There now, have a coffee, and go design for the Web, and everything will be OK.

Mezzoblue has some great tips about debugging your CSS. These are all practices I’ve followed in my own CSS debugging and are worth bookmarking.

He also raises a good point about skills you should be looking for in a modern Web designer:

…the most valuable skill to possess in the maddeningly complex minefield of today’s browser landscape isn’t, in fact, knowing which browsers do what to which properties. It’s problem-solving.

Listen to the man, he knows what he’s saying.

One final, Mezzoblue related, note. For a project at work (soon to be launched, I promise), I used Mezzoblue’s RSS tip to create an XSLT sheet to transform the RSS into HTML for users who inadvertently click on the link (expecting something other than XML code). Seeing an XML file transformed to styled HTML before your eyes is pretty damn cool.

Posted in Web design | Comments Off on Un-fixing your layout, fixing your CSS