The book Peopleware has this to say about quality:
... managers tend to think of quality as just another attribute of the product, something that may be supplied in varying degrees according to the needs of the marketplace. It’s like the chocolate sauce you pour onto a homemade sundae: more for people who want more, and less for people who want less.
The builders’ view of quality, on the other hand, is very different. Since their self-esteem is strongly tied to the quality of the product, they tend to impose quality standards of their own. The minimum that will satisfy them is more or less the best quality they have achieved in the past.
My experience of the way I feel when working on a software product fits well with this description.
Peopleware further says that:
Quality, far beyond that required by the end user, is a means to higher productivity.
I include the second quote just to show that it is a potentially defensible position that high quality software will not put a software shop out of business, even though the opposite view seems to be oddly prevalent. Arguing this point is not the goal of this question, however.
I would also like to present some examples of other questions on this site that indicate that engineers caring about the quality of what they work on is not an uncommon occurrence:
- Professionalism: protecting code quality
- My work can affect peoples safety, and my employer prefers rapid development over quality?
- How can I fulfill my personal quality-demanding in a low-quality environment [closed]
- Perception of quality by management [closed]
With that background out of the way, my question is how can I identify that a software company provides an environment where quality software development work can be done? I am interested in effective methods for determining this at any/all stages from finding a job ad to the end of the interview process, inclusive.
Here's some ideas I was able to come up with myself, along with reasons I don't think they are sufficient, or necessarily even predictive at all:
Looking for descriptions of quality work being valued in the job description and/or the company's website. I think it's quite easy for the word quality to end up on the marketing material without anything in particular to back it up. One can imagine someone finding "quality" as a synonym for "good" in a thesaurus.
Asking about software quality during the initial phone screen. Unless the company is very small, in terms of employee count, I would not expect the person doing the phone screen to have any idea about this. I would guess they would say something like "of course!", or perhaps pass that on as a question to someone else. Asking about that here may also affect the chances of getting to the next stage of the interview process.
Directly asking whether quality work is allowed/encouraged during a technical interview. While this interview stage has the greatest chance of being with someone who would know the answer to the question, those same people have the most motivation to say something besides that answer! Some reasons I can think of include: They have not faced the uncomfortable truth of the quality of the code. They may be simply embarrassed about it. Finally, they may just have only ever seen low quality code, and so may think they work at a reasonably high quality place, despite that not being the case. Also, many technical interviews do not leave much room for questions like this.
Directly asking whether quality work is allowed/encouraged during a follow up interview, if one happens. Since these only happen sometimes, it doesn't make sense to solely rely on them as a diagnostic. Even if one was willing to rely on that, these meetings are often with folks who work far away from the actual software development, even if they were closer on the past.
Picking up on subtle context clues during the previously described stages. This seems like the best way in this list, but it seems far from accurate to me.
Are there other options that I have not considered here? Are one or more of the options that I listed above more effective than I currently think?And just to get ahead of some likely comments, this is a highly unbalanced binary classification problem, but I'm looking for something more accurate than the sadly quite accurate baseline of "assume quality is not permitted".