View Full Version : how do developers test unstable software

16.09.2005, 12:45
i was wanderin how do software developers test unstable versions of their software. for eg. the new version of daemon tools was written from scratch, so in the beginning it would be unstable, and if used might mess us the system. do developers use software that can emulate the windows environment or sumthing else. and what is the windows kernal and why is it so important to software development. i'm just a bit confused :oops:

16.09.2005, 13:21
we use several methods to test our software (emulators, tests-scripts etc.,
virtual machines)

One of the most important parts is for sure beta-test, where we send out
the beta-versions to different persons/systems. Those betatesters risk
to damage their system and must check for all kind of different installations
and possibilities of configuration-probs etc. All in all it's a very time-consuming
process, because we always find scenarios that nobody thought off before.

After that several customers, who cooperate with us, receive ReleaseCandidates (usual customers with big network/system-installations,
even several hundred machines) to check stability.

Only if we're pretty sure that we iron out the bugs we're aware of (you never
can guarantee for the absence of bugs, you can only guarantee that there
are always bugs (we even found newer ones in V3.47) and minimize them.

When all parts run stable on every test-system, we release it to the public