Webslug: Helping measure the worldwide wait… sorta

Webslug claims it can “help you measure the worldwide wait.” Cool tagline, but I’m skeptical.

The time I got back on one site was infinitesimally small (less than a twentieth of a second) when personal experience shows it takes around 10 seconds. Another site showed over 40 seconds when my experience has been around 20 seconds. And finally the site doesn’t answer the question, “How fast is it at a given connection speed?” It just gives seconds to load without the context of a data rate. (The Firefox Throttle Add-On does this nicely, but only on the Windows version.)

On the upside, it does let you compare two sites to each other and collects measurements of all the sites that get tested for public review. I like these ideas, but Webslug will have to fix those deficiencies before I come back often.

One final wishlist item, the value of which was demonstrated in a presentation by Aaron Barker recently, is being able to see the difference between “page appears” (something happens to show the user that the site is loading), “page is partially functional”, and “page is fully functional” (100% loaded). If a page takes too long to appear initially, the user may get impatient and hit Refresh—which will actually delay them further by restarting the whole process. So even if it’s not fully functional there is value to getting something up there quick, even if it’s not the full meal deal. But I’m not sure an automated tool could really do that… Someone prove me wrong please!

See Webslug for yourself.

Also check Yahoo’s best practices for increasing performance on the front end. (Thanks for all your recent help on this stuff Aaron!)

posted by Ted Boren on Thursday, Jul 10, 2008
tagged with tools, performance, connection-speed