Tag Archives: internet-explorer


Microsoft’s just launched modern.IE – a collection of tools which aims to make the task of  testing web pages and applications on IE easier.

The main tools of interest are an online page scanner (no good if your page isn’t public), and virtual machine images (now for some additional platforms, including VirtualBox) for browser/OS combinations going back to IE7 on Vista.

Also came across the IE10 compat inspector script, and a good post about best practice when using vendor prefixes via the new site.

IE10 will drop support for conditional comments

IE10 will be dropping support for conditional comments. As stated in the post, feature and behaviour detection is preferable to browser detection – but with so many quirks in previous versions of IE, those aren’t often enough. So conditional comments to add classes to the html element for IE<10 will stay around for quite a while longer.

Empty image src can destroy your site

I recently investigated a bug which caused a page load in IE7/8 to make two requests to the server, the second one without a querystring for some mandatory parameters – which caused some warnings in the server log.

Turns out that image tags with an empty src attribute cause this to happen in at least IE7 and 8. So if you have such a thing on a common page elements such as a header or footer, the load on your servers is doubled – not good.

There’s an in-depth discussion of this over at NCZ online – Empty image src can destroy your site, and also a solution using the base tag to prevent this from happening in IE.

The best solution of course is not to have such tags in the first place, but it’s good to have a defence given how easy it is to never notice the extra request.