| Re: [ide-dev] Survey results |
Remember you asked for my opinion. J IMHO, surveys every 2-3 weeks is a really bad idea. Making product decisions based on a the majority of a simple survey is a really bad idea. Doing a survey with lots of questions about individual compiler warnings is a questionable idea. Survey fatigue is a well-known fact. If you do too many you end up getting fewer responses and only those people that like doing surveys. For example the Eclipse community survey that we run on an annual basis would easily get over 1000 respondents but now it is very hard to get over 700. You also have to remember that a survey is just ONE data point. It should be used to show a trend or sentiment, not an absolute decision. There are lots of factors in making product decisions so saying you will make a decision based on the simple majority of a survey results is just not good product planning. Finally, doing a survey that asked people to answer yes or no for each compiler warning would be very tedious to fill out so the results would be suspect. The longer and more difficult the survey the more skipped answers you get. I hope this helps. Remember you asked my opinion. J Ian From: ide-dev-bounces@xxxxxxxxxxx [mailto:ide-dev-bounces@xxxxxxxxxxx] On Behalf Of Mickael Istria On 12/10/2013 06:06 PM, Lars Vogel wrote:
I don't know, I have no experience with surveys. @Ian: what's your opinion on this?
If some warnings have bugs, then we simply won't propose to enable them (bugs must remain hidden). They should even be removed from the Preferences page IMO. -- |