Tips and Tricks Archive
Must every app nowadays have "tags?"
November 22, 2005
The answer unfortunately is "yes." It is a requirement. But what are "tags?" What are they used for? And how do they help me in Test Run?
To be perfectly honest, "tags" are nothing more than keywords, although if you listen to a lot of innovators today they may make you believe that they are a new invention. What is new about these keywords called "tags" is not the concept itself, but how people are using them. Tags are being used today by communities to organize content. Whereas a single person may build an "ontology" using categories and keywords, a community builds a "folksonomy."
One thing that is different about how applications are using tags is that they are made visible to the public. Typically keywords were considered private metadata associated with an asset, and only a small minority of people could view or edit a set of keywords. Keywords were used to help facilitate searches for assets within a catalog and contained words and phrases that may not be contained in other metadata fields associated with the asset.
Tags on the other hand are made visible to everyone and (in some systems) editable by everyone as well. This may seem like a scary idea at first and one horribly prone to user error and malicious "tag spammers." But experience across the Internet with tags has shown quite the opposite, that a community is quite capable of policing, managing and editing quite a large number of tags and assets.
Also, not only are tags visible, they are often hyperlinked. Often when you click on a tag you will subsequently be displayed a page containing assets that share that tag. For each of those assets you will see another set of tags, which are also hyperlinked. In the end, tags not only provide a categorization scheme for content owners, but a navigation scheme for visitors as well.
How does Test Run use tags?"
In the first iteration of Test Run I relied on a highly structured schema for a test case. I needed a way to track:
- test case type (e.g. functional, regression, unit, usability, etc)
- product area or categories (e.g. "UI Front End," "Search," "File Uploads," etc)
- keywords
- software version and build numbers
I even had users submitting feature requests for me to add even more fields to a test case. It was simply not scaling, the UI was getting strained, and there was too much field bloat.
So I decided to reduce all of those fields to a single one: tags. In Test Run tags allow the user to label a test case in an unlimited number of ways. If you associate a rich set of tags with all the test cases you create, then when it comes time to generate a test plan you simply say, "give me a test plan with test cases that have been tagged 'registration' and 'functional' and 'email verification'." And presto, a test plan is generated and is pre-populated with relevant test cases.
In Test Run tags are used to keep things simple and flexible at the same time. Tags provide structure, but allow users to define their own structure as opposed to being reliant on an arbitrary structure I impose.
The bottom line is that there is nothing to fear. Use tags the way you would use associate keywords in any other content management system or application and you will soon be addicted.
Enhance Your Test Plan Spreadsheets
November 20, 2005
Test Run does not force you to completely change the way you conduct your quality assurance process, it simply allows you to execute it more efficiently. One way it achieves that is by allowing your team to leverage the tools that they are already familiar with, like Microsoft Excel or another spreadsheet application.
About Byrne Reese

Byrne Reese is a product manager by day and an engineer by night.
He conceived of Test Run to help project managers like him stay up to date and informed of what his team was working on.