Fenders and Benders

The tech team at Ai is split into two parts. Roughly two-thirds of our staff are developers, the behind-the-scenes programmers and creators of software and rich applications.

The other third are front-end engineers, handling the HTML, CSS, and scripting languages rendered by browsers. Both teams are tight-knit and collaborative, particularly the front-end team.

In this spirit the smaller crew coined itself a little ways back: they are the Fenders, short for "front-enders" (obviously). They take great pride in their work, compete for compliancy accuracy, and play some mean foosball. Most importantly, they work as a team. With a great name.

By extension, the developers are Benders, for back-end, although the term hasn't made the same impact. The Fenders, on the other hand--or "Fendas," as our Bronx-style lead Fender likes to say--are really making a name for themselves. Two of our clients have started using it regularly.

The next time you hear about web page creation, don't think simply in terms of client-side coding or web design. Think: fender.

Wither Windows?

So quite recently, the Ruby on Rails open source web framework announced that they would be migrating from the Subversion code repository they had to a new one managed by Git.  Git is a version control system created by Linus Torvalds to manage the Linux kernel.  Linus had several requirements in mind when he made Git, requirements that involved specific sets of features, scaleability, stability etc.

As might be expected from the creator of the Linux kernel, none of these requirements included running well on Windows.

Git does technically run on Windows, but its kind of a hack, and Redmond's favorite platform is definitely treated like a second class citizen (ooh... irony...).  So naturally when Rails moved to Git, there was a number of Windows users who were concerned they were being left behind.  Interestingly, the Rails maintainers responded that amongst the core developers of Ruby on Rails, Windows users were a small minority.

So then, in this other piece I was reading (I need to see at least two things before I declare an Official Trend) John Dvorak rips on Dell, claiming they're stuck in a 90's mentality.  In the article, he says Dell isn't keeping up and startups in Silicon Valley these days tend to use laptops, and many many of these laptops are Macs.

Even Senator Schmelkin, a long time Windows guy, switched completely over to a Mac a couple of months ago (I tried to get him to blog it...sorry, no luck...).

Okay - I knew Apple was getting a boost from the whole  iPod thing, but I never expected to see quite this level of momentum (and yes, yes...I'm sure in the accounting and parking facility businesses Windows still has 18456% market share...).  There seems to be an accelerating trend, especially in the software and web world where not only is it more desirable to work on a Mac, but its beginning to look like people are beginning to take the position that Windows doesn't matter.  It's like it's deprecated.

(Disclosure - I was a Mac guy from before it was cool, except for a span of about 5 years that I spent trying to install Linux on a laptop).

The Rails guys do tend to be a bit religious at times - "my way or the highway".  But I do find the basis for their switch interesting.  The lack of first-class support for Windows was simply not a consideration.  Has the world finally changed?  Is the wicked witch finally dead?

They're baaack!

My life has been previously sliced up into distinct, and mostly non-overlapping chapters. I've lived in Toronto, Montreal, San Francisco and New York (and a couple of adjacent suburbs that I won't get into). In each place there have been a number of people that I've known. In time, I've moved on and only marginally keep in touch with friends from the old 'hood. I'm what they used to call a lousy letter-writer.

It's not so bad. I have great memories from each place, and my life is organized into distinct chunks. I've gotten used to the idea of thinking in terms of  "so-in-so from my former life at location X". Frozen in time, these memories provide a nice back drop for my identity.

Now, however, Facebook is screwing it all up. My entire life (in people) since high school has all been sucked into Facebook. Everyone I've ever known is already there. And they're all friending me (or I them). Twenty years of time, compressed into nothing.

When yet another mysterious figure from my past "friends" me, I'm suddenly privy to some very disturbing photographs, where I can see the effects of 10 to 20 years of aging all at once. ("Dude, you have no hair! What happened?") Often they're accompanied by mysterious young short people in these photographs. Short people that weren't there before.

I feel like I've fallen into a time warp, or maybe am watching my life flash before my eyes. Except it's the life that I didn't live, all the paths that I didn't go down, with people that I didn't keep in touch with. They're all back. They're on Facebook, they're my friends and they've escaped from their neatly delimited chapters in the past to invade my present.

Search Design

How do you find the information you're looking for on the Internet? This, of course is the original reason the Web was created - as a publishing platform for the physicists at CERN.

As the web matured, people constructed websites with tree-like structures, viewable as a site map. As web sites grew in size and complexity, we started referring to a large amount of information organized in a tree structure as a taxonomy. Some people (I'm looking at you, Yahoo) tried to taxonomize the entire web. Here's what we found out about taxonomies: they don't scale. Once they get too big, they start to collapse under their own weight. Historically it was a mark of a professional that they could navigate a large taxonomy, such as the Dewey Decimal System, or Scientific Classification. I don't believe that we'll see many new taxonomies of that scale adopted by the world at large.

So that leads us to search. Search, of course, is the other method we have to find the information we're looking for, especially when all we want to do is get to the information, and we don't give a rat's hiney how it's catalogued. But since the average search length on Google is something like 1.3 words, we need to infer a tremendous amount of information from very little input.

To me, this seems like a tremendously deep opportunity for the developers of websites to shape how information is presented via the establishment of business rules modeled within search. I call this Search Design, and I see it as a critical part of how websites with large volumes of information should be built.

Search design is currently implemented primarily in terms of how information on a website is indexed. This indexing controls the relative importance of information, and the relative importance of different structures inside the information, and is generally implemented by the developers who built the website.

However, at a higher level, search design is something that can be expressed as a series of business rules. Say you have a newspaper site with articles and authors. You could establish the following business rules for search:

• Articles have higher priority than Authors
• Weight of an article should be, in order of priority: Article Title, Article Author, Article Tags, Article Body.
• Weight of an author should be: Author Name, Author Department, Author Tags, Author Bio

It could be a lot more sophisticated than this of course, it could contain relative numeric values for the indexed values, and the business rules surrounding relative importance of information could also be far more sophisticated.

Search scales. It can handle a LOT of information, but as implemented on individual websites, the presentation of it to date has been relatively naive. Search design could grow to be a profession in itself, much in the same way that information architecture developed over the last decade. For many sites it is search, not a taxonomy structured navigation, that is the primary means of navigation for a website.

Pricing right

A fascinating study was released last week that observed people find expensive items more desirable simply because they're more expensive. The same bottle of wine is more appealing to consumers who believe it's worth $90 than others who are told it's worth $10.

Pricing reactions also rely on contextual cues. Panasonic has found midprice items sell better when shown next to pricier models. The "reference prices" around a product significantly affect their perceived value.

These discoveries are not really new. Consider the make-up artist my wife hired for our wedding five years ago. After agreeing on a price, he forgot what they had decided, and quoted a second rate 80% higher. When my wife expressed shock, he quickly reverted to the original price. He explained that many of his clients were upper-class women who wouldn't think he was as good a stylist at the rate we were paying. Raising prices actually appeals to his audience, the cost verifying the talent on offer.

These theories are particularly interesting with regard to online merchants. As noted in the New Yorker article, websites create price transparency, which makes for savvier consumers both online and off, and forces retailers to compete in other ways.

The most intriguing question is raised by reading the two articles above as a single unit. In the first, The Economist argues that prices create emotional response, which could be used to sellers' advantage. In the second, James Surowiecki suggests almost the opposite: that the Internet has eliminated much pricing opacity, creating empowered consumers who understand right-priced items. Observing which one proves more accurate--or whether these theories work in parallel--will help define a generation of sales strategies.