I wanted to post this as a reply to the blog post “I know you'd have a test case for this !!!” by Ajay, but since the reply grew long, I posted it on my blog.
I agree with James Bach stating that this is a platform issue. It’s very clear that the issue reported here is more to do with the platform than the application itself.
This makes me wonder why Windows XP was chosen to run such an application. There could have been many reasons like ease of compatibility, familiarity, db connectivity, interoperability, ease of development, etc. However one needs to keep in mind that Win XP is not designed to just run in the background, it is an end-user interface for applications. So, Win XP should have been customized to meet the needs of deployment.
Now let me try answering the questions posted in the blog.
- Is it a bug? How risky is it to ignore such messages?
- If I go by the definition used by James and Michael “A bug is something that bugs somebody who matters” – yes I would call it a bug.
- How risky could this be
- Well, if there is a 24/7 support answering the travellers queries I do not feel it is risky, since the affected traveller could get their query answered, also knowing this balloon from Win XP, it hides after 10s or so, I do not feel it a very risky bug.
- Is the purpose of the display served?
- Yes, the purpose of display is definitely served, if there is a pop-up marginally blocking the display for a few seconds, I would not go to the extent of questioning the display. I agree that there could be % of customers who might be annoyed, but definitely the display is served.
- If the overlapping of the message on the display is a bug, will you fix it?
- To me this is a redundant question apart from “will you fix it” part. Will I fix it – since this issue could be resolved by disabling the informative balloons in Win XP, I would fix it. The cost to fix it very low, so I guess it could be a straight forward decision.
- What if it is not fixed?
- If it’s not fixed, it might continue to annoy a % of users.
- Which tester will think of these kinds of tests?
- It’s interesting how everything and anything boils down to a tester, may be because he is the one who touches the product last before it’s shipped. I feel issues like this could be minimized when the engineers (dev & testers & deployment) have enough domain knowledge or aware of the deployment or know their exact mission. This check might be of highest priority in the checklist for engineers from SCADA or HMI domain (Well, they might have a test case for this:).
- Do you wear the hat of a non-tester and say: Hmmm, there is a workaround. I'll not fix it.
- Who is a non-tester in this context?
- I do not know how to fix it.
- Are you kidding me? :)
- It might be a bug but it is a limitation of the technology. [Cannot fix]
- WHT? I seriously have no words to say.
On further scouting for similar information over the web, I came across this very informative link on how to turn Win XP to a HMI platform. In the article author writes about different parameters such as performance, security, disabling un-wanted balloons to make Win XP suitable for HMI platform.