Microsoft has decided to withdraw one of the patches it released this month after complaints from users who couldn’t open PowerPoint.
The patch – KB2920732 – was meant to be a boon rather than a bane but failed to deliver its promised aim of stability, mostly with video playback. Instead, many users started to complain about issues.
According to Windows Central, a pretty big problem arose straight after the patch was installed – PowerPoint would no longer open on devices running Windows RT.
Commenter BGrewer explained the problem he was having:
After performing a system update on my Surface RT (first gen.) I found that I was unable to open Powerpoint, even though Word and Excel still work. I have shut down and restarted the system several times and Powerpoint still won't open.
Screenshots of error messages uploaded by BGrewer reveal PowerPoint was reporting how it was “either not designed to run on Windows or it contains an error”.
Clicking on OK and attempting to run PowerPoint again prompted an error message which offered to run the program in safe mode. Clicking Yes on that brought no joy – Windows would instead attempt to reinstall PowerPoint without success.
Fortunately, Microsoft is aware of the problem and is currently working on a fix:
Shortly after the release of KB2920732 , Microsoft became aware of an issue affecting users of PowerPoint 2013 on Windows RT devices. We have removed KB2920732 from the Microsoft Download Center and Microsoft Update and will provide a new update as soon as a fix is available.
In the meantime, Windows RT users can refresh their devices by following the instructions on Microsoft’s Support site.
Microsoft notes that the refresh will keep personal files and settings intact but also points out that if the refresh fails, the only remaining option may be to reset the device – which would itself of course lead to a loss of data (make sure you backup any important files first).
Microsoft appears to be having a hard time with its patching system right now with a string of failures over the past six months.
In August we wrote about MS14-045, which had to be pulled after causing the dreaded Blue Screen of Death.
Then, in October, it had to pull an update designed to add the SHA-2 hashing algorithm to Windows 7 and Server 2008 R2, saying it was “investigating behavior associated with this update”.
In December the run of patch failures accelerated – three had to ultimately be withdrawn. One prevented Outlook connecting to Exchange 2010, another locked out further updates on machines running Windows 7 Service Pack 1 (SP1) and Windows Server 2008 R2 SP1, and a third – MS14-082 – caused issues with ActiveX forms and broken macros when Office 2013 and Office 2007 were run side by side.
While Microsoft generally acts quickly to address borked patches, the amount we’ve seen recently may leave some home and business users wondering whether they should take their chances on a manually delayed or critical patch update-only schedule.
And that is neither a safe nor desirable position for Microsoft customers to find themselves in.
Image of safety pins courtesy of Shutterstock.
Mark
Looks Like you missed KB3013455 which messes up your fonts seen on vista and win7 Had to uninstall it here!
Dick Holman
Migrating to a Unix-type OS would negate these problems.
Paul Ducklin
I doubt it would *negate* these problems, because update-related incompatibilities are in no way exclusive to Windows. Apple has had this sort of problem many times before; I can think of any number of Linux OS and app updates I’ve applied in my time that caused me problems. (Drivers that stopped working, apps that would no longer work, old files that could no longer be loaded, etc.)
You might mitigate, or reduce, or feel more in control of the process…but I am 100% sceptical of your choice of the word “negate.” It sounds like the sort of denialism that used to lead Linux and Mac users to declare that because there weren’t many viruses for their platforms, it therefore followed logically that there weren’t, in fact, any viruses. And after that faulty leap of logic, the next inference was that there weren’t any viruses because viruses were actually impossible.
Counter-example for you: iOS 8.0.1, which is very much a “Unix-type OS”…the update actually broke mobile network connectivity for some people. You can argue that’s a lot worse that PPT files not opening on Windows RT :-)
https://nakedsecurity.sophos.com/2014/09/26/apple-pulls-ios-8-0-1-update-tells-users-to-roll-back-to-ios-8/
MikeP_UK
Proper UAT would have found these problems before release. But Microsoft and other software vendors do not do UAT at all! They only run scripted testing based in scripts developed by the software development team and are only designed to find errors on the coding against their understanding of what the code should do. So it misses completely the checks about whether it does what was intended or not, whether it actually works in a ‘real time’ environment, does what users want it to do in the way they expect it to work, etc, etc. Only UAT and real user style testing does any of that – scripted testing does not. Hence all the problems. Please can we have properly tested software again?
Before retiring I did UAT, real user style testing and wrote user manuals and training on actual software rather than plans that can so easily get changed.)