Since the May 12th, when there were 50 or so respondents, another 50 people took part in the logback-DSL survey bringing the total participants to a little over 100. The new results are Do not care for either 37.9% Groovy 29.1% Scala 28.2% Other 4.9% I find the stark changes in the results quite fascinating because over a uniform population survey results should converge very quickly, say after 50 or respondents, so that adding new participants should not significantly influence the outcome, at least only by a few percentage points at most. So either the population of respondents was not uniform, in the sense that those participating the beginning were somehow different than those participating later. Alternatively, the web-survey system we are using is vulnerable to manipulation so that few users can unduly influence the survey's outcome. On 12/05/2010 7:57 PM, Ceki Gülcü wrote:
Hello all,
As you probably know already, there is an ongoing survey about the preferred language for the logback DSL. See [1].
Here are the results so far.
Do not care for either 25% Groovy 25% Scala 50%
Interestingly enough, most respondents express a strong preference for Scala over Groovy even if Groovy seems to have a larger user base.
Cheers,
-- Ceki