It had 40 upvotes and 4 years has passed
It has 40 000 views. Anyone considering to start learning java will google "future of java" All they get is this outdated thread.
The problem is, its not a good question.
It is a poll of opinions and is attracting poor answers
I would say it's on a decline. It's not gone, but it's past its peak.
It was losing importance in the mobile market, but with Android, it has regained it again.
Not sure about JAVA. But surely the JVM will remain relevant supporting many other languages.
These are not good answers. The question is attracting such answers and that makes it a not good question either. Good questions attract good answers. Poor questions attract allow poor answers.
The quesiton has 28 answers, including some that have been deleted:
The Magic 8-ball says:
Outlook not so good.
It's been historically locked because it is an historical curiosity of the site - it does not represent a good, high quality question on the site. Furthermore, it cannot be 'fixed' to be a good one (its flawed and has answers that trying to transform it into a good answer would mean deleting most of the answers). The other option would be to delete the question.
The number of views doesn't mean that it is on topic. It means its a popular search and one that was done in the earlier days of the site when questions and answers were more care free.
That it shows as a high scoring result on Google is a testament to Stack Exchanges SEO work and the quality of the site as a whole. It does not speak to the appropriateness of an individual question in the Q&A format.
We are programmers, software architects, software engineers, and developers (and whatever other job title that people heave made up for our skill set) with the occasional CTO tossed in too. What we are not is futurists or prognosticators. There are very few people who have any academic or professional training in futurology on the site... and it's not within our skill set as expert programmers.
That point, we have opinions, and can have discussions about what the next trend will be and what will be a fad and gone tomorrow (seriously, Cobol and Fortran had their day... no one writes code in those languages anymore... err... wait... they do). And while you may want to know our opinions, a polling of opinions is not a good question in for the Q&A format. This is why the original question was closed.
If you want to know today, what the outlook for Java in 5 years will be, go ask a career advisor. Ask your professor if you should be learning Java (or for that matter, if Java is what they are actually trying to teach - or if they are trying to teach other concepts). Ask futurologists for how to prepare for the next technical revolution (or if we're in a bit of stagnation for a period of time).
Asking us, however, isn't going to give you the answer you want. You will get every answer from "its days are over" to "it will love longer than Cobol" at which point, well, you haven't really learned anything.
To the point of "should I take a class in Java?" It depends on what you are trying to learn and what it is going to teach you. If you are going to be learning "this is how to do XYZ in Java", no. If you are going to learn "this is how to do XYZ, and Java is our implementation language", yes... and well, thats an opinion. You need to make sure that you are learning how to learn a new programming language. For me, in college, I had classes that used a total of nine different programming languages (nearly every class used a different one (one class used four)) and I can assure you that while I would be challenged to do a spline in fortran 90 today... or much of fortran 90 for that matter, you give me another programming language to learn and I can pick it up fairly quickly.
And so... back to your question:
You do happen to have 20 rep on the network. You may find that asking in The Whiteboard chat room you can get a discussion... though I will tell you now that you will get a wide range of answers there too (I'm a javaist, though there are some ceesharpers there too who disagree with me about the long term outlook of different technologies).