Just this morning I was talking to someone about UTest’s newsletter where we agreed that the newsletter was just an advertisement (the context was AST newsletter). Unlike AST, they are a commercial organization and all their actions are to generate more revenue.
Somehow I got directed to UTest’s blog through some other website where one of their guys blogged about learning to testers from Boeing787 disaster. It was a funny read because this blog post did not make any sense to me. Obviously, it was part of their marketing campaign. But my problem is, I like to see advertisements that make sense. This guy was simply trying to compare apples with, ummm, I assume carrots. Read the blog post:
I responded to the blogger, but apparently they are not accepting the responses anymore. May be they are aware that it will get bashed. So, I decided to post my response to Mr. Roskill and one commentator here. Read on:
Dear Damian Roskill and Rui Alves,
Both of you are talking and agreeing about what happened to Boeing 787 and what could be learned and improved by Boeing etc. Can I ask you a very basic question please? How much you know about aircraft manufacturing and testing process? From looking at your profiles I can say that both of you have no background in aviation industry.
Damian, you asked, “What does this have to do with software testing?” I’d say, nothing. Because you are comparing apples with oranges. The battery incident has nothing to do with software.
Then you say, “And software, in most cases, isn’t a matter of life or death.”
I agree, software is not a matter of life & death; but not in ‘most’ cases, rather ‘many’ cases. Surfing web or using your phone software will not kill you as long as you are not doing that standing in the middle of a freeway (highway). However, what should also be noted that software applications are used on a large part of our daily transportation systems and today many vehicles are controlled electronically through their computerized systems. Also, you do not seem to be aware of regulatory aspects of aviation, medical & military software processes apart from other transport systems where regulation is not as important. The software used in these specific areas may become a matter of life & death.
Then you say, “Boeing is a case study in another way: nothing you do in the confines of the lab can prepare you for the real world.”
This also shows how naive you are. Aircrafts as well as their software do not get tested just in labs, they are tested in the sky, including covering scenarios representing load factor similar to real passengers.
Before FAA approves airworthiness of an aircraft, it has to go through months of test flights called operations approval phase. If FAA (or any other aviation authority for that matter) thinks that a software or hardware change requires a certification, that change has to go through a Supplemental Type Certification (STC) process.
Rui: Just for your awareness, when we develop & test software that will be used ‘on the aircrafts’, pilots are involved in all phases of software development. There, software testers too test more than test requirements & positive testing.
As they say, little knowledge is dangerous thing!