William Stanek here. My blog posts to Microsoft Press are sometimes delayed, as is the case here with my blog post about a mobile code bug that was driving me absolutely bonkers. The code bug was this:
Often when you visited web sites and were using Internet Explorer 8 on 64-bit Windows 7, you were redirected to a web page optimized for mobile browsers—and mobile browser pages aren't exactly what I call user-friendly. Most mobile browser pages provide only summary information, which is nothing like what you actually searching for.
I gave as examples:
- www.etrade.com, which redirected to wireless.etrade.com
- www.intel.com, which redirected to m.intel.com
- www.sony.com, which redirected to m.sony.com
Because the post was delayed by nearly a week and because I already had made three other blog posts about it for O’Reilly, the issue was largely resolved by the time my post appeared on the Microsoft blog and this is why readers who took a look at the three example sites I provided didn’t see anything.
To be clear, I first blogged about this on 12/31, then again on 1/1 and again on 1/4:
- IE8 64-bit Windows 7 and the Mobile Code Bug. http://bit.ly/7OPrpy
- IE8 64-bit Windows 7 and the Mobile Code Bug (Part 2). http://ow.ly/Tqph
- IE8 64-bit Windows 7 and the Mobile Code Bug (Part 3). http://ow.ly/SG7j
The bug existed whether you were running 32-bit or 64-bit IE8 on 64-bit Windows 7. I first noticed the bug after what I assume were a series of code changes on affected sites, and I assume the bug had to do with the interpretation of the USER_AGENT field provided by the browser. As an example, Sony Corporation of America made changes to their site, www.sony.com, in early November. On Nov 11, I reported to them that there was a problem, specifically I told them the following:
When you enter www.sony.com using Internet Explorer 8, you are redirected to your web site for mobile phone browsers http://m.sony.com/. This is a problem other sites are experiencing because of HTML/coding compatibility issues with IE8.
My error report should have been more clearly worded. I should have stated that this was a problem specific to the way site code redirected browsing for IE8 running on 64-bit Windows 7. Why? Well, no action was taken when I reported this error to Sony (and ditto with others previously). I assume what happened is that the webmaster or whoever believed all was okay. After all, they probably visited their company’s site all the time with IE8, though likely not with IE8 running on 64-bit Windows 7.
The good news is that word spread quickly after my initial post. I saw changes by the next day and by Jan 4, 24 out of 25 sites I bookmarked as well as the 3 example sites were all good.
The important lesson here is this: 64-bit platform testing is every bit as important as 32-bit platform testing.
Web designers, webmasters, coders, etc., need to perform appropriate testing on both 32-bit implementations and 64-bit implementations. They need to determine which platforms, combinations/variations are relevant/in focus and which aren’t. Clearly, you can’t test every variation. However, in a world rapidly shifting to 64-bit platforms from 32-bit platforms, thorough testing of 64-bit is as essential as 32-bit testing. That’s the important lesson learned because if heavy weights like Intel, Sony, and Etrade can make this mistake, so can anyone else.
Thanks for reading, time for me to get back to work! Hope you'll take a look at my new book Windows 7: The Definitive Guide. Also just released is my book Exchange Server 2010 Administrator's Pocket Consultant.
William R. Stanek
williamstanek at aol dot com