Talk:Bug (engineering)

Definition

I'm guessing I'm opening a can of worms, but ...

WRT the 1st sentence: "In engineering, a bug is a defect in the design, manufacture or operation of machinery, circuitry, electronics, hardware, or software that produces undesired results or impedes operation"

I like that it is rather boldly worded. That's good. but I think it somewhat confusing and ambiguous and worse is off the mark.

In what sense can a bug be a defect in operation of something? A bug causes an issue during operation. It's not a defect in operation. Maybe in here means during which grammatically it can. But, what is a defect during operation? Something can be used incorrectly; user error. But, that's not called a defect; it's a mistake.

I don't think a bug can be in manufacture either. Certainly, defects occur during manufacturing; manufacturing defect. But, that's not called a bug, is it?

Consider in the design. I agree that a bug is a defect in the design. Further, it is a defect in both the process of design and in the design as an artifact. Design has those two aspects, but manufacture and operation don't. They are only processes. Therefore, I find the result of combining all of these terms into one sentence to be ambiguous and confusing.

IMO, a bug is a systemic issue. It occurs during design and becomes part of the design. For software, that may seem wrong since programmers separate design from coding, but coding is in some respects design as well; a different level than other software design. But, in particular, coding is definitely not manufacture or operation.

No, the bug in that notebook is not about a systemic issue. But that was a joke. I've always thought that story was funny and interesting; as most do. I've also always thought it was silly since a physical bug is not what bug means in an engineering sense. As far as I can tell, it never has. It is intriguing and humorous to visualize something that's not physical as something that is.

Moving on: rather than trying to list all types of engineered systems just call it what it is. I admit that system is not a great word, but I can't think of a better noun. I do think that engineered is spot on. I think that a bug can only be in an engineered system; in a thing built by a human. This is a critical, defining aspect of the term. A defect in a natural system would never be called a bug. For example, humans have faults but not bugs.

Lastly, I think impeding operation is generally an undesired result; one example of zillions. Well, if someone finds an operation being impeded to be desirable, then it's not a bug :) So, what's important is whether someone finds the behavior desirable. The definition does not depend on their personal preference; only that they have one.

So, what is a bug? Here's my suggestion: A bug is a defect in an engineered system that causes an undesired result.

What do you think? Stevebroshar (talk) 14:01, 28 April 2024 (UTC)

Move duplicate info from hardware and software bug articles

IMO this article should for the most part supersede both software bug and hardware bug. I think hardware and software are no less and no more than classifications of bugs. There are zillions of ways to classify a bug. These two are pretty special since one can argue all bugs are either hardware or software. Not sure whether hardware and software should have their own articles. But at least for now I do suggest that duplicate info from those pages get eliminated; moved to this article. In particular, I think the history section is duplicate in all three although I'm sure not exact dups so merging should be done carefully to not lose info. ... I'll do it, just wondering if anyone has concerns about my plan. Stevebroshar (talk) 15:05, 28 April 2024 (UTC)