Module Insights

What it is and why it exists

Before a review, a baseline or a status meeting, someone always asks the same questions: how many requirements are still in draft, who owns them, how many have a safety classification, how many are not linked to anything yet. The answers are in the module, but counting them means filtering, scrolling and adding up by hand, or building a report in a reporting tool, which takes time and often needs an administrator.

Module Insights answers those questions inside the ELM Mini Dashboard next to the module. It reads the module once and draws charts of the values of any attribute and of the link coverage. Every slice and every bar lists the requirements behind it. The widget only reads; nothing in DOORS Next changes.

Problems it solves

• No quick overview of how a module is distributed: status, priority, owner, artifact type, classification, dates.
• Requirements without links are found late, often only during a review or an audit.
• Counting values by hand with filters is slow and the numbers are out of date the next day.
• A report in a reporting tool has to be built, maintained and shared before it answers a simple question.
• A number alone does not help: the question is always which requirements it stands for.

Benefits

• Answers in seconds: open the module, draw the charts and read the distribution. There is no report to prepare.
• Any attribute: enumerations such as status or priority, yes/no attributes, dates by month, the people who created or own the requirements, the artifact types.
• Link coverage: how many requirements are linked for the link types and direction you choose, and how many are not, split by link type if you want.
• From a chart to the requirements: click a slice or a bar and the requirements are listed, with a jump into the module for a single one or for the whole list.
• Ready for a meeting: the charts and all their numbers go to Excel, with the requirements behind every value.
• Safe for everyone: the widget only reads, so anyone may use it at any time, also in a baseline.

How to use

1. Open the module and look at the charts. Each chart is a grey block that says what it counts: the values of an attribute, or link coverage. Change opens its settings: a Name for the chart, the attribute or the link types and direction, an Artifact type to count only those, and a View: donut, pie, bars, columns, lollipop or treemap. Done folds them again.
2. Click Draw charts at the bottom. The widget keeps the charts in this browser, reads the module once and fills every chart; a progress bar shows how far it is and you can stop it at any time. Next time the widget opens, the saved charts are built at once for the open module; with nothing changed the tile says Read again. Clear saved charts deletes them again.
3. Add chart puts another chart under the others, up to six; the × removes one. Click Draw charts to read and draw it.
4. Click a slice, a bar or a row of the table: the requirements behind that value are listed under the chart. Click a row for its details with Show in module and Open in DOORS Next; Highlight in module selects the whole list.
5. The image icon saves a chart, or all charts at once, as a PNG or JPG picture. Log saves every value of every chart to Excel, with a sheet per chart listing the requirements behind it.

Good to know

• Requirements without a value for the attribute are shown as (empty), often the most interesting slice.
• A chart colours up to eight values; rarer ones are gathered in Other, which can be clicked like any other slice. The Excel file lists each of them.
• If an attribute can hold several values at once, a requirement counts for each of its values, so the shares can add up to more than 100 %. The chart says so.
• Link coverage counts a requirement as linked when it has at least one link of the chosen types and direction. Links to other IBM ELM applications are not counted.
• Whether a link is marked as suspect is not available to widgets, so Module Insights does not show it.
• The charts never show requirement texts, only names and identifiers, and the widget changes nothing: it can be used in a stream, a change set or a baseline.
• Very large modules can be done in parts: stop the run, reload the page and continue where it stopped.