6.9.4 Installability measures

What this heading covers

Installation is a countable event and a common source of support cost. The measure should include the clean environment, the time, the number of steps, the failures and the rollback. An installation guide is not evidence. A product that installs only after manual dependency work has a lower installability than one that installs from a package in one command, and the measure should reflect that difference. Uninstallation and reinstallation belong in the same family because a partial uninstall breaks the next install.

Record position: family 34 of 40 in clause 6, under 6.9 Flexibility, which carries 4 families. It is one of the 40 families a programme has to map its measures to, and its elements are listed below. The heading is 6.9.4 in the public contents; the text behind it is not public.

How a measure for this family is built

Count installations that completed without manual intervention as the numerator over the attempts. Report the median time and the number of steps, and count failures by cause. Include the uninstall and reinstall cycle, because it is where leftover state shows up.

What to put in place before publication

1. Start from a clean environment every time.
2. Include an uninstall and a reinstall in the test.

Evidence to keep

Two questions to ask

• Was the environment clean before the install?
• Was an uninstall and a reinstall part of the test?

Where teams usually fall short

• An install measured on a machine that already had the runtime and the dependencies.

What the number cannot tell you

It cannot tell you whether the installation is secure. A one-command install can still open a port it should not.

What happens next

Replaceability is the family for swapping the product for another, which is the question an acquirer actually asks.