-
Notifications
You must be signed in to change notification settings - Fork 3
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Principles for web developer research #1
Comments
Outlined this in the doc about joint research, adding it here for context: To ensure research efforts done together across multiple browser vendors and organizations, this document aims for agreement on general principles and approaches regarding research: Principles
Benefits
|
Filing #2, a variation of "Web developer time is precious" springs out. For surveys that web developer volunteer to do, it makes a huge difference if they survey is fun and/or educational. IMHO, State of JS/CSS are very strong on the fun/educational aspects. |
I've created Principles for joint developer research One area that still needs to be discussed is
|
The web platform supports a lot of different use cases. Different people use it in different ways. I'd probably want to avoid focusing on a definition of who a web developer is that is too narrow. In fact, even the name "web developer" might not necessarily include all users of the web platform. My feeling is that we should learn from as many folks as possible in this research, but open to hearing if there's a need to focus on a given type of web platform users rather than others. |
I think the question of who we want to learn from is intimately tied to the question of what we will do with the results. In general, I think we will want to use it to help prioritize improvements to the platform and its developer experience; but there are many ways to prioritize among for instance:
I'm also guessing that depending on the research, we may target different parts of these, and different subsets (e.g. developers using CSS vs JS). Some research benefits from breadth (as many respondents as possible), others may benefit a lot more from depth (why are so few web sites adopting technology X). So maybe we shouldn't answer the question in the abstract (who is a web developer), but it's probably worth starting to explore what kind of research would need what kind of input, and then dive into the principles that drive audience identification across these different kinds. |
When conducting research and running surveys, what are the most important things to get right? What are common pitfalls we should avoid?
A few things that come to mind for me:
The text was updated successfully, but these errors were encountered: