AJAX Heresy: It's not a feature and nobody really cares anyway

Lately I've been hacking around with AJAX. Most of the time AJAX is used for little more than eye candy but I thought I had found a situation where AJAX would do actually something novel and solve an existing problem.

The problem is on our universal whois lookup site: easyWhois.com, where it has sporadically run into problems over the years where one registrar site or another (*cough* cough Netsol *cough*), would block us for "excessive queries". The idea was to use AJAX to push the query back into the client's browser, instead of doing it from easywhois. The server would be more of a switching station: it would grab the domain to be queried from the browser, find the appropriate whois server for it, send that back to the client who then would connect to that server and conduct the query. Nice, easy, simple and has the effect of difusing all the whois queries across the client IPs instead of at the central server, vastly reducing, if not eliminating blockages at remote registrar whois servers.

So it literally took 30 seconds to get Rasmus [Lerdorf's] 30 Second Ajax Tutorial working and begin to understand what all the hoopla was about. A couple of other Ajax examples later (here and here) and the entire idea took a hit. The idea in it's original form won't work because the javascript XMLHttpRequest only lets me do HTTP GET and POST method connections (as the name would imply). There doesn't seem to be anything like PHP's fsockopen() available to me to get the client to open its connection to port 43 (the whois port). I'd have to use something like Java, and if I did that I may as well load the entire application in there, never mind Ajax.

Throughout this process something became clear to me which I think is perhaps lost on a few people, Ajax is a tool, period. Granted that it is an essential component of "the Web 2.0 Kool-Aid", it is not a feature and I don't see it as an end-user selling point. It became even clearer as I explained to my wife (who is far more technically adept than the average person) what Ajax was and did and watched her eyes glaze over. And it wasn't an incomprehension glaze over, it was a "who cares?" glaze over.

"Ajax allows one to create bi-directional communications from the client to the server without page refreshes"

0.0001% of the population: Wow!

Everybody else: *yawn*....what's a "page refresh"?

So in the course of all this, and what inspired this post, was MyLongTail.com which is self-described as "The first AJAX based system for Search Engine Optimization". Something I came across co-incidently during my foray into Ajax and which hammered my point home for me. MyLongtail seems to be along the same vein as MyBlogLog.com, in that they are both search engine keyword trackers and IMHO, offer a subset of something like Indextools, which uses the same approach of embedding javascript into the remote pages to collect data and has been around for years.

But I digress, so MyLongTail talks up it's Ajax-ness. If you look at their signup form, you really see a good application of Ajax in form validation. This in itself makes Ajax worth the price of admission because I can see this being a very effective means of providing less cumbersome form validation and probably drastically reduces form abandonment. Who wouldn't want that? Answer: nobody. Roughly the same number of people that would actually care that the technology that enables this is something called Asynchronous JavaScript and XML, a.k.a "AJAX".

In short, I don't see the point in trying to turn your tools into features. We use all kinds of various tools to provide our services and most of our customers couldn't care less what they are. They just want the box to go "Bing!" when they press the "Bing!" button and the rest is our problem.

It is beyond the scope of 99.9% of web users to even be aware of what Ajax is, and of the remainder, less care.