The results of this years Eclipse Open Source Developer Survey include some interesting finding about what motivates participation in open source projects and motivates developers to build apps in their free time.
Compared to previous years, the survey noted that corporate policies are becoming more positive towards open source participation with only 0.6% of respondents choosing the "Does not allow the use of any open source software" to describe their organizations policy.
By contrast the most popular response (39.5%) indicated both using and contributing to open source projects and another 9.7% were in organizations that made significant contributions to open source projects:
A new question in the survey investigated why individuals participate in open source projects. With only 28.5% admitting "I don't contribute to any open source project", 71.5 of developers surveyed participated in open source projects, with the altruistic response "Like to give back ..." being the most popular choice (54%):
Another question asked whether respondents developed software/applications not related to their day job.
Ian Skerrett, Director of Marketing for the Eclipse Foundation who is responsible for the survey commented:
I was a bit surprised that 84% claimed to spend some amount of personal time developing software. The key reason is to learn new technologies, 74% answered they ‘enjoy programming and learning new technologies’ and 71% ‘keep my skills sharp’. An important lesson for anyone in the software industry that is targeting developers: Make it easy for developer to learn your technology.
The survey was conducted from April 23 to May 15, 2012. Of 840 individuals responded, 732 completed the survey and the finding are based on 732 completed survey responses. 52% of respondents have over 10 years experience writing code in a professional capacity, 52% have 2-10 years experience writing code and 12% less than 2 years. Overall 90% are satisfied or very satisfied with Eclipse:
Trends indicated by this years survey are that more people are making plans for cloud computing and 21% have already deployed an application to the cloud with Amazon Cloud Services being the most popular at 36% although the use of private clouds has increased to 30%.
Mobile computing is now more firmly established with only 24% having no plans for mobile apps, a decrease from 32% in 2011. Android is the most popular target (89.6%), followed by iOS. (73.8%) and the rest trail well behind - Blackberry 11.4%, Windows Phone 11.1%, Symbian 2.3%, Bada 1.3% with 19.9% choosing "We optimize our website for mobile devices". With regard to the use of mobile frameworks and platforms, Skerrett comments:
It is a bit surprising that more developers are not using cross platform frameworks. 60% claim to use only the Mobile OS SDK. jQuery Mobile (28.6%) and PhoneGap (17.9%) are the most popular mobile frameworks.
Other findings concern operating system, language and IDE preferences.
While Windows is still dominant (55%) this is a decrease of 8% from 2011; Linux showed a 4.5% increase to 32.50% and Mac OSX increased by 3.5% to 12%.
Java remains by far the most popular language with 76% naming it as their primary language
Apache Tomcat continues to be the most popular application server with 34% of respondents using it for deployed applications and Spring continues to be the most popular framework for those doing server side development (23%).
Although Subversion is the dominant source code management system its popularity again decreased (now 45% down from 51% in 2011) and it is Git (23%) that is gaining ground. Ant was the clear leader as build management tool with 51% share but Maven had increased in popularity (from 13% in 2011 to 27%) overtaking Jenkins and Hudson both of which fell back.
As we predicted over the weekend, when we noticed a web ad for it that had been shown prematurely, the Raspberry Pi Model A+ has been launched today. The additional news is that it will sell for [ ... ]
We have recently seen how much neural networks can learn, but the latest example is very strange. Google has succeeded in teaching a neural network to be a computer simply by showing it programs and w [ ... ]