Ajax Performance

WebWait

Michael Mahemoff, of ajaxpatterns.org and author of the seminal Ajax Design Patterns*, has released WebWait, an Ajax-aware, in-browser performance testing tool. You load a URL within an iframe (with sexy lightbox opacity, naturally), and the total page load time, including background Ajax calls, is reported in the parent window (and a sexy lightbox overlay, naturally).

It’s a neat tool because the testing runs within your Firefox, Opera, and IE browsers and includes all of your local cookie settings. There’s a flip side to that coin, though, as caching effects are going to obscure most content download contribution to page load time. I’m also a bit concerned that there isn’t a tremendous amount of room for the tool to grow; currently you only get start time to page onload level granularity out of the captive iframe, and I’m not sure it’s possible to collect much more than that with this technique. XSS protections make it impossible, as far as I know, to collect DOMReady or more granular load time metrics, but I wouldn’t be surprised if Michael has some tricks up his sleeves to overcome those limitations.

As I’ve posted previously, collecting Ajax performance information directly within the browser is compelling and provides a level of insight that simply isn’t available with other techniques. It’s great to see so much focus on Ajax-aware performance testing these days.

* Michael’s book was a freebie for attending Ajax Experience Boston, but I never received my copy due to some sad twist of fate involving inadequate or late shipments, hint-hint if anyone from Ajaxian happens to be reading

Catching up on the news bin

January was a brutal month for work and play, so I haven’t been around here as often as I would like. I’ve also accumulated a bit of a backlog of small news items. So, let’s churn through those:

• Local storage is all the buzz. Ajaxian carried a brief summary of the latest news in the space yesterday, including a very detailed article by Niall Kennedy explaining the technique and the upside. Meanwhile, the irrepressible Brad Neuberg continues realizing that potential with Dojo Offline.
• In other Dojo news, Alex Russell posted a brief gripe about the incredible slowness of regular expressions, especially for replacements. The gripe included a link to a very detailed survey of regex suckage on multiple platforms, for those who want the deep dive.
• Contrarian Peter Michaux thinks there is still a window.onload problem, despite previous declarations of victory by Dean Edwards. By way of background, Peter’s article begins with the need for a cross-browser DOMContentLoaded event before embarking on a historical tour of attempted solutions. Finally, Peter arrives at the increasingly popular DOM polling technique as the best solution. Great article.
• If you’ve read many of my articles here, you know that I’m a big believer in perceived performance. Often times in a complex app, you are constrained by technical factors out of your control (a slow, aging database, for example), but you can still improve user experience by hiding that latency behind careful design decisions. On the flip side, as this editorial by Dion of Ajaxian shows, it’s even easier to make poor design decisions that kill perceived performance.