A recently completed Java.net poll indicates that, excepting one big forecasting mistake, 2013 turned out pretty much aps Java/JVM developers expected, when it came to their own work. This past January, a poll asked developers to predict which category of new Java/JVM technology developments will affect their own work most in 2013. The just-completed poll tested the accuracy of those predictions.
A total of 226 votes were cast in the year-end poll. The exact polling prompt and results were:
My work in 2013 was affected most by new developments related to...
- 6% (13 votes) - The non-Java-EE Cloud
- 20% (46 votes) - Java EE (in or out of the Cloud)
- 20% (46 votes) - Java SE / OpenJDK
- 9% (21 votes) - JavaFX
- 13% (29 votes) - Java Mobile (including Android)
- 1% (3 votes) - Java Embedded
- 7% (16 votes) - Non-Java JVM languages
- 6% (13 votes) - Java Tools (IDEs, CI, etc.)
- 4% (8 votes) - Other
- 14% (31 votes) - My work wasn't much affected by new developments in 2013
Looking at this, there is no single predominate area that most affected Java/JVM developers' work this past year. But, you can say that developments related to Java EE, Java SE, and Java Mobile had the greatest impact (53% combined) on what developers were working on. I think this is good news for Java EE in particular: in part it suggests that the release of Java EE 7 back in June is looking like a success.
So, this is the data for what actually affected developer's work. Now let's compare this with what the developers predicted would affect their work back in January 2013. That poll asked developers to respond to: "In 2013, my own work will be most affected by new developments related to...". Here are the comparative results:
|The non-Java-EE Cloud||53%||6%|
|Java EE (in or out of the Cloud)||8%||20%|
|Java SE / OpenJDK||10%||20%|
|Java Mobile (including Android)||10%||13%|
|Non-Java JVM languages||2%||7%|
|Java Tools (IDEs, CI, etc.)||2%||6%|
At first glance, the match-up doesn't look that great. This is due to a clear mistake in the January forecast: back then, a majority of developers expected their work in 2013 to be affected most by the non-Java-EE cloud. Yet, as we bring 2013 to a close, this actually happened for only 1 in every 17 Java/JVM developers[standard caveat: Java.net polls are not scientific polls!].
If we set this forecasting "mistake" aside, a different picture emerges. Back in January, after the non-Java-EE cloud, developers expected new developments in Java EE, Java SE, and Java Mobile to have the greatest effect on their work in 2013. Lo and behold, in December they reported that, indeed, new developments in Java EE, Java SE, and Java Mobile did indeed have the greatest effect on their 2013 work. Not a bad job of forecasting, in my opinion!
In fact, if we apply an ordinals system of looking at the results (the Olympics are coming up soon!), the January forecast looks incredibly perfect, if you eliminate the "non-Java-EE Cloud" mistake. Here are the ordinal results for all of the technology categories except the "non-Java-EE Cloud" and "Other":
|Java SE / OpenJDK||1 (tie)||1 (tie)|
|Java Mobile (including Android)||1 (tie)||3|
|Java EE (in or out of the Cloud)||3||1 (tie)|
|Non-Java JVM languages||5 (tie)||5|
|Java Tools (IDEs, CI, etc.)||5 (tie)||6|
Now do you see what I mean in saying that Java/JVM developers did quite well in their January forecast of what new technology developments would most affect their work in 2013?
New poll: conferences and post-conference online materials
Our current poll asks Does the eventual availability of technical conference materials online affect the value of attending the conferences in person?. Voting will be open until Friday, January 10.
Subscriptions and Archives: You can subscribe to this blog using the java.net Editor's Blog Feed. You can also subscribe to the Java Today RSS feedand the java.net blogs feed. To follow Java.net net on Twitter, follow @javanetbuzz.