Download Database

First, the simplest: the DOWNLOAD database button.

When clicked, this button will download a copy of the current database (by default, called “database.sqlite”). The app developers recommend doing this anytime you plan to make integral changes to the database (like changing metric weights), before anything else.

Updating Weights

The CONFIRM button allows you to view existing and change metric weights used to calculate a risk score using riskmemtric.

Here you can either select a metric from the Select metric dropdown box or select a row
in the “CURRENT RISK SCORE WEIGHTS BY METRIC” table.

Once the metric is selected, enter a new (numeric) weight in the “Choose new weight” box and click the CONFIRM button. You’ll notice the row with the change will appear highlighted, showing you the old weight value and the new weight value:

After you have updated at least one metric weight, the UPDATE button is enabled, allowing you to apply new weights and re-calculated the risk for each package.

After you’ve made all final re-weighting changes, you can click the “Update” button pictured here. Note that these weight changes will take effect on all packages in the database, updating the risk score for each.

Once you click on UPDATE you will get a confirmation modal, reminding you that “updating the risk metrics cannot be reverted”, and strongly recommending you download a copy of the current database first before clicking on SUBMIT. The modal also details all the actions that will take effect in the app if you submit these changes. Note that starting with Version 1.0.0, this will be a re-calculation on the stored {riskmetric} assessments, not a re-assessment.

Once you click on SUBMIT a modal dialog box will appear indicating that it is applying weights and updating risk scores to each package you have uploaded. This will be followed by a brief “updates completed” message.

Now go to the database overview tab to confirm that the risk scores have been updated. You can also view the new weights, including the standardized weight on the Assessment Criteria tab.

return to top

Select Category Colors

To change the color for one of your decision categories, select a new color by either typing in the desired hexidecimal code or use the color input.

To apply these changes, click APPLY COLORS. A pop-up will ask the user to review and confirm the selection:

The modal contains a summary of all decision categories with their respective new colors. If you are happy with your selections, simply click SUBMIT and the database and application will be updated with your selections.

Automate Decision by Risk Score

If there are no decision rules applied, you’ll see the following decision categories adjacent empty check boxes. When setting up these rules for your organization, you can choose as much or as little automation as desired.

For example, if you want to set up an automatic decision for “High Risk” scores, you only need to select that check box and make your risk score designation / cut off value. From then on, every time a package is uploaded with a risk score > 0.67, it will automatically receive the decision label: “High Risk” which saves your review team time and mental capacity when reviewing large groups of packages.

Similarly, if you have package risk thresholds mapped out for each decision category, then apply them here! The user interface won’t allow the slider ranges to overlap, but it will allow gaps between each range. In the example below, we adopt automatic decisions for packages with risk scores less than 0.33 and greater than .67. In this fictitious example, let’s say our organization has a policy that any package whose risk score falls between 0.33 and 0.67 will need to be reviewed manually.

To apply these changes, click APPLY DECISION RULES. A pop-up will ask the user to review and confirm the selected automation rules:

Upon clicking submit, the automation rules will be stored for all users to see on the ‘Upload Package’ tab. After uploading a new set of packages, you can see the ‘Summary of Uploaded Package(s)’ now includes a summary of decisions made and each category it was automatically assigned. In the example below, you can see that 10 packages were uploaded, and 7 automatic decisions were made: six received the “low risk” designation and one was labeled as “high risk”. In the summary table, you can note the risk score and decision on the right-hand side. You can also observe that a few packages weren’t found because they don’t exist on CRAN.

In the ‘PACKAGE CONTROL PANEL’, if we select one of the packages where a decision was made, you will see that the ‘Status’ has been changed to “Reviewed” and the risk decision slider has been adjusted accordingly and disabled. In addition, an ‘Overall Comment’ was submitted, stating that the decision made was driven via the automation rules.

return to top