6

I've been wavering back and forth about this site. I want to contribute, but on any given day, I tend to see a lot of super basic "How do I click a button with Selenium" questions and not a lot of interesting SQA theory and practice questions. Should we do something about this? Can we introduce some guidelines about what types of questions we want here?

Examples of questions that bore me, going down the recent questions list:

Not all Selenium questions are bad, obviously, but 22 of the 50 newest questions are about Selenium, and many of them are that basic. I'd like to see more interesting questions about Selenium if that's what we're going to be spending almost half our questions on, you know?

7

Here's an answer based on my usual hangout, cooking.SE:

Create a canonical question

Cooking.se has a lot of people stop by and ask "Is this meal still good, it's been sitting on the counter for four days and turned green" (only minor exaggeration there). They created a canonical question: "How can I tell if food is still safe to eat", with a comprehensive guide to what spoils when. Any new question about food going bad gets closed as a duplicate to that question; the asker gets their answer, and the community as a whole isn't stuck answering the same basic question over and over.

We could easily create a few basic "How do I X in Selenium" or a "Getting started with Selenium" and rapidly close any question whose answer is found there. Three of the questions I linked to as boring are the same basic question, "How do dropdowns work". They would all be closed as duplicates (and honestly should even if we don't create a canonical answer). If it's a basic reference for how to start using Selenium, all but that one dynamic button question would be a duplicate.

I moved discussion of this topic and progress to: Creating canonical references for Selenium basics

  • I actually like both of your suggestions, but I'm voting for this as historically, I don't see a lot of downvoting happening in SQA, and it needs a critical mass in order for it to work. But this could work with just a handful of people willing to create good canonical questions and mark up duplicates. It also creates useful content for the site - win-win to my mind. – testerab Dec 29 '14 at 22:24
  • Ok well, over a week later, I'm not seeing much interest in general, so let's start doing this. Do you have suggestions for what canonical questions we should make, @testerab ? – Yamikuronue Jan 7 '15 at 16:38
  • I think your list gives a good starter - "How do dropdowns work?" would be a good question to start with - there are dozens of questions on dropdowns, so a good solid question and answer would be a good resource. – testerab Jan 7 '15 at 18:27
  • 1
    Wow, it's been a busy 20 days. Anyway, I started with "How to get started", writing a comprehensive answer and making a few small edits to the question to make it more generic. sqa.stackexchange.com/questions/11785/… – Yamikuronue Jan 27 '15 at 17:06
  • This is a good idea and I feel we should also have a canonical question for "Different methods to upload files using Selenium Webdriver". I have seen the "upload" type of questiion too many times on Stackoverflow, don't know about SQA. – LittlePanda Apr 15 '15 at 10:58
  • @littlepanda good idea, please add it to the linked question where I am collecting ideas for canonical questions :) – Yamikuronue Apr 15 '15 at 11:37
5

Here's an idea based on Puzzling.SE's ongoing issues:

Downvote more

Super basic questions that show little to no research effort should be downvoted. It's right there in the tooltip: "This question does not show any research effort; it is unclear or not useful". If we band together and downvote bad questions, it shows what kind of questions we want and which ones we don't. Puzzling.SE (an even newer community than this one if I remember rightly) is having a similar problem: too many poorly-thought-out challenge riddles with not enough effort put in to make them any fun. They're aggressively downvoting anything that is vague and undefined, and it's increasing the quality of their questions.

3

While I agree completely with your concern, my fear is that the basic questions from users is what leads to better questions in the long term if they get the assistance they need.

My initial reason for visiting was a question regarding Selenium IDE, since then I have become a somewhat active member.

How do we, as a community, create better quality and more interesting questions?

I think the first thing we should do is set up Community Wikis on the basics of QA and Selenium. Then we close questions as they come up that would be covered by the Wikis.

  • I like this, it's basically what I was getting at with the canonical question idea – Yamikuronue Dec 30 '14 at 16:07
  • Yeah, I was going to add more to it but about to head to lunch. Essentially, the more experienced users also need to start asking more questions, not just answering. – Paul Muir Dec 30 '14 at 16:08
3

Two points. First, before this site was launched, I believe there was a proposal for an independent Selenium site. That was eventually merged into the proposal for the site you're reading now. The net result is that this is where a lot of people go on SE for Selenium questions.

Second, the majority of Selenium questions are naive questions because the majority of Selenium users do not know a lot of programming. People with experience in manual testing look to automation for further their careers and frequently, for better or worse, the first automation tool these people investigate is Selenium. Often these folks have no programming experience whatsoever. Sometimes they even think Java is a feature of Selenium. Given their circumstances, it is understandable that their questions seem so naive.

I'm not sure that voting to close Selenium questions that are boring to you is going to encourage more interesting Selenium questions. If you want questions to be more interesting, write some yourself.

2

I think the problem is where to draw the line of saying "this is bad" and "this is interesting" as it will surely be individual by different users. Also I see part of "the problem" as selenium is much about scripting like a programming langauge and not a behavior like "how can I improve my TDD".

If the boards meaning they don't want "selenium coding" questions there's only one solution: close and ban selenium questions. But if we (and I do) want to talk about selenium here as it's a common test tool than we have to handle these questions the same as we handle beginners questions on SO e.g.

  • 1
    Really? That's all you've got? We're doomed to be SO, swamped by basic questions? Let me throw out some alternatives, see if we like any of those. – Yamikuronue Dec 29 '14 at 15:20
  • 3
    It's entirely possible for us to decide that we don't want "You could google for this in 3 seconds" questions. And improving bad questions and/or voting to close them is something that the SQA community can do to improve the quality of what we see on SQA. – testerab Dec 29 '14 at 22:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .