The other day I had an annoying problem accessing SharePoint sites. I am running Windows 7, IE8 and Office 2007 and Office 2010 Beta.
Any internal SharePoint site continued to throw the error of element not found for init.js
It only happened on my desktop (my laptop worked just fine). I tried clearing the temporary internet files, browser cache, even uninstalling and reinstalling IE8.
Nothing worked, I finally found the solution here – http://social.technet.microsoft.com/Forums/en-US/sharepointgeneral/thread/41e1874f-7825-4bfe-ba4c-37255d198df2
So, I opened up SharePoint Designer 2007, ran the diagnostic tool in the Help menu, restarted IE and everything was back to working as normal.
As discovered by Todd Carter there is a memory leak in SharePoint 2007. He has outlined the details as well as a work around to the fix the memory leak.
I decided to do some testing with it, however, when I compiled the dll, placed it into the GAC, and changed my global.asax file, I started getting this error: “Could not load file or assembly ‘[Assembly Name]’ or one of its dependencies. The system cannot find the file specified.”
After digging into this error for a while I discovered my problem. This may be fairly obvious to developers out there 
, but coming at this from more of a SharePoint administrator type roll, I missed this one step that isn’t outlined in the steps provided by Todd.
After deploying your dll to the GAC, you need to open up your web.config and place
<add assembly="[Assembly Name], Version=[Version Number], Culture=neutral, PublicKeyToken=[Public Key Token]" />
in the web.config file of the web application you wish to apply your fix to. This line should be added between <assemblies> and </assemblies> in the web.config file.
All of the information: Assembly Name, Version Number and the Public Key Token can be found by right clicking on your dll in the GAC and viewing the properties of the dll.
Once I added the assembly to my web.config for the web applications I was trying to apply the fix to, everything worked as expected.
I had a client email me with an interesting dilemma a couple days ago. He was working with a document library and uploading .docx files. He was creating a group of Office 2007 files and Office 2003 files and uploading multiple files at once. He emailed me saying that all the files were uploading but .docx files. Sure enough, they weren’t uploading and no errors were being generated.
He emailed me a screenshot of the upload, and I noticed something strange. The size of the Word document was 0 K. I created my own blank Word document, saved it and uploaded it and it was 10 K. Curious about the difference, I thought he had a corrupted or “bad” Word file. Nope, it open just fine in Word, but was still 0 K.
Come to find out, he was right clicking on his desktop and selecting “New Microsoft Word Document” I was going into Word and saving the open file. The file created by right clicking on the desktop, is a valid Word file, but is missing some key info that was causing the upload to fail.
If you open a .docx file in WinRAR or WinZip, you’ll notice it is actually a collection of folders and XML files. When saving a blank Word file from within Office, there are some default XML files that are created and added, however, when you create the file by right clicking your desktop, none of those XML files are created. If you try to open such a file in WinRAR you get an invalid archive error as opposed to the folder and file structure you usually see within a .docx file.
The lesson, if you’re going to do testing with a blank Word document, create it from within Word and not by right clicking on your desktop.