Which kinds of documentation Arduino developers use to define the system requirements? Which testing tools and levels of tests do they use?

Projects Discussion and Showcase Education and Teaching
1

Greetings,

I am student Sara Guimarães from the Graduate Program in Systems and Computing (PPgSC) at the Federal University of Rio Grande do Norte (UFRN) and I am conducting research on Arduino documentation and testing. I would very much like to count on your cooperation by answering a brief form of questions. The form is short and only takes a few minutes to complete.

Follow the link to access the form: https://forms.gle/QLFxjf8jFYHKkUAHA

Thank you very much for your collaboration. Any questions, please do not hesitate to contact us (sguimaraaes@gmail.com).

2

no one will click on a unknown link.

4

Do you think they would access the link below? Both point to a form on Google Forms.

Link: Which kinds of documentation Arduino developers use to define the system requirements? Which testing tools and levels of tests do they use?

5

A lot of Arduino software is written by amateurs and ex-professionals who are unaware or pretty tired of the whole ISO9000 processes that leave software meeting some "letter of the law" requirements WRT testing and documentation that don't seem to actually success in making it more usable to bug-free for the end users. You write the code, try it out, make it seem to work for some number of trials, and then publish it on github with a README and hope that other people find it useful, or at least submit problems in a form that helps you make it better.

The fact that most things are open source helps a lot.

Speaking for myself, anyway. Hopefully Arduino (the company, with people actually getting paid) is a bit more formal. But I don't think i've ever seen anything resembling the 'functional spec', 'design spec', 'test plan', etc that I used to have to write. (For OSSW, you'd think they would be published as well, right? Hmm.)

6

These parts of the system design are essential and also expensive.

7

In my 33 years of programming and data processing, they are always wrong! Even when they cost a fortune and done by professional consulting firms.

8

Yes, you are right.
These extra costs mostly due to uncontrolled system changes and a none existing change management.

9

guimaraaes needs to come to the conclusion that ALL the things she has listed are only created to get a contract signed. They will only become pertinent if the project fails and the lawyers get involved. This is the usual case when dealing with government agencies where nether the agency nor the vendor understand what is to be produced.

10

I wouldn’t go THAT far. As an engineer and manager, I’ve found the distinction between “functional spec” and “design spec” to be very useful, even if the FS is “just implement the damned standard.” More so if you’re trying to implement some kind of overreaching “architecture.”

Test plans are much harder, but there are people who are much better at it than I am. I have a particular pet peeve against “must pass regression tests” requirements when your legacy feature don’t HAVE regression tests, and no one is budgeted to write them…
Alas, testing gets much more difficult when you have a very broad set of platforms and applications.

Even in the arduino forums, we’ve seen the value of “code reviews” (and some of the problems with them, as well.)

11

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.