Quote

"Between stimulus and response there is a space. In that space is our power to choose our response.
In our response lies our growth and freedom"


“The only way to discover the limits of the possible is to go beyond them into the impossible.”


Thursday, 3 November 2016

The Conundrum of Automation Frameworks



While the industry is realizing the importance and criticality of a proper test automation practice in delivering world class software, but still it is very rarely looked as a ‘profit center’. The moment ‘cost center’ word gets associated with a practice all the ‘eagles’ of the world decent to look at ‘measures’ to ‘save’ cost and run with ‘bare minimum’. The ‘required focus’ to ‘criticality’ ratio still remains miserably poor. To augment the lack of ‘desired focus’ it is often tried to be supplemented with the ‘available’ and sometimes the ‘available and not so useful’. The result: an inconsequential, mediocre, fragile, maintenance intensive, unreliable automation practice. It barely does the job for an abysmally low percentage of targeted functions which results in low usability and low confidence of stake holders. Under tremendous delivery pressure and customer escalations there is no choice but to organically boost the manual testing team, work overtime etc... etc…. The story goes on and on, release after release.  

The problem is not only on the planning and focus front but also on the available or appropriate talent pool front. While manual resourcing challenges are a different ball game which need to be addressed with the right approach there are readily available technical solutions in market and community which if employed with right customization can turn out to be the perfect fit for the insurmountable in-house problem since ages.   

‘Agile’, dependable, extensible, cost effective, rapid in-sprint test automation is very difficult to achieve without having a proper extensible automation framework in place. Most teams rely on internal development/automation teams for development, enhancement, and customization of frameworks backing test automation efforts. However the framework is a complete product in itself which requires commensurate product development focus. Again because of the cost and possibility of the product to fail, most orgs shy away from setting up such a framework development team. So if we do not want to develop then the only options are to look at the market or go to the community. As a community and or practice matures it is likely to churn out some reliable solutions which can be customized and adopted.

But if you go out hunting in the market you need to be very clear of what your ‘exact’ need or requirement is. Unless you are clear with your requirements you are highly likely to pick up a misfit or a solution for another problem which you do not have to solve.

So what is the requirement? Let us take the high level requirement as ‘‘Agile’, dependable, extensible, cost effective, rapid in-sprint test automation’.
So how to achieve this? A framework or a solution which is ‘easy to use’, ‘does not require huge learning curve’, ‘easy to write tests as in English-words/KWs’, ‘handy reporting and logging’, ‘platform or application agnostic’, multiple interfaces (API, Command, GUI), data-driven, decoupled test assets, and integration with CI. If you are planning to build, keep these features in mind. Even if you are not planning to build there are other solutions…

Since we have setup the boundary requirements, now we can proceeed to the next steps...  I will try to touch other aspects and solutions in upcoming series…


To be continued….  Stay tuned... J