>> If an end-user application crashes during normal use, it is ultimately the >> fault of its software developer. > > Or user error.
If user error can lead to a crash, that crash is ultimately the fault of the software developer. > Or memory failure. If the crash is caused by running out of memory, then it's not the fault of the user, but of the developer who either neglected to test whether an allocation succeeded, or even if he did so test, failed to handle failure gracefully. If the problem is a bad memory module, then the fault lies with the electrical engineer who designed it, or with the manufacturer. The user would only be to blame if he handled or installed the memory improperly, say by not observing antistatic procedure. > Or processor overheating. Then the fault would lie with the engineer who designed the computer, unless the user blocked the ventilation holes. > Or disk corruption. That's likely the fault of the software developer who wrote the filesystem code. > Or (Windows) registry corruption. That's microsoft's fault for making such a critical system component so vulnerable to corruption. > Or earlier power outage. It is not well-understood that journaling filesystems only prevent the filesystem from corruption. Filesystem journals don't protect end-user data. In the case of OOo, the documents are compressed zip archives of small directory heirarchies. While I expect that makes it easy to create, read and write the documents, the document format is especially prone to corruption in the event of power failure, application crashes or system crashes. > Or malware. The vulnerability of Windows to malware is largely Microsoft's fault. OpenBSD has a history of very, very few vulnerabilities. I've never heard of an exploit "in the wild" attacking OpenBSD. >> Perhaps you have a friend who already has a Linux or OS X box that you >> could test with. If it works OK on Linux, then you could install it in a >> virtual machine such as VirtualBox. > > > This may well be a useful test, but if the problem doesn't show on the > friend's system, it may still be any of the above. The change of operating > system may be a red herring. True, but my objective in suggesting this was to help the OP present his presentation to his intended audience. Finding the cause of the crash can come later. --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@openoffice.apache.org For additional commands, e-mail: users-h...@openoffice.apache.org
