Sudhi

Posted on May 4, 2019May 4, 2019 by Sudhi

Unconventional ways to improve testing skills

There are many conventional ways to improve testing skills like understanding different software testing techniques. How about unconventional testing skills?

1. Play spot the difference and memory recall games

2. Find #defects while watching a movie – Remember best of the directors make mistakes

3. Learn to #code and think like a #developer – Sometimes thinking like a developer helps in identifying where developers make mistakes

4. #Questioning – Why, How, Where, When, Which, What…

5. Meditation – Definitely improves concentration

6. Sleep properly – A good night’s sleep is always helpful

7. Listen to music – Relax and rejuvenate

8. Dancing – This is experimental. Just kidding 🙂 Instead automate your tests or checks to get more time to improve and test your #testing skills

9. Join the #Developers, Open Source, Linux, AI/ML/DL… group with 87,000+ members using https://lnkd.in/evv4bt9

10. Join the Software Testing & #QA group with 210,000+ members using http://bit.ly/2N3uvU5 or follow the group on Twitter @standqa

Note: I’ve to admit some of them are conventional!!!

Posted on April 26, 2019April 26, 2019 by Sudhi

AI vs Common Sense

Do we need an AI system to determine the fields where AI will be very useful?

Artificial Intelligence is making inroads into every field. But do we need AI in each and every field?

Everyone understands the fact that AI will be very useful in certain fields where it will take generations to solve the problems. But there are certain fields where aggressive usage of AI could potentially create more problems.

Do we need to use common sense or build another AI system that would forecast whether AI would help or create more problems? Can we call such a system as Artificial Common Sense???

#artificialintelligence

Posted on April 12, 2019April 26, 2019 by Sudhi

Should testers worry about Automation or Artificial Intelligence?

I’ve seen many #testers who are reluctant to add automation to their bag of skillset. I can broadly classify them into two categories

1. Testers with exceptional #testing skills – These Testers have the ability to test an application at an amazing speed. They either adopt automation or simply don’t care about automation.

2. Testers who are afraid of automation because of the speed at which it executes the tests or checks, whatever term you prefer. They often compare automation with manual testing. These Testers perhaps think automation will be able to think in future.

Can automated scripts think?

1. An automated script written to test a scenario can just validate a test or check.

2. Smart automated scripts can take decisions. These scripts are written to cover a story rather than a scenario. This is not intelligence, but creates an illusion.

3. The other solution is #ArtificialIntelligence.

Testers need to worry about human intelligence because #AI is trying to emulate what humans exhibit and not what automation exhibits.

But Yes…Artificial Intelligence being intelligent would either remove testing by developing perfect #code or self test the code by adopting #automation!

#softwaretesting

Posted on April 9, 2019April 26, 2019 by Sudhi

Can AI define requirements???

Recently someone stated on social media there is a solution to test automation problem using AI. Suggested solution is to read manual test cases using AI and generate the automated scripts.

My reply to that solution:

1. If AI has the ability to read manual test cases, AI can definitely read user stories.

2. If AI has the ability to develop automated scripts by reading manual test cases, AI can definitely code by reading user stories.

3. If AI can read user stories, develop software and automate, it can definitely generate test cases.

The final question that remains unanswered is whether AI can generate an idea or define requirements???

Let’s say there is a way for artificial intelligence to generate an idea or define requirements, there are lot of questions that need answers. The first question is Why should AI generate an idea or define requirements?

If you have an answer, please leave your comments.