What Are Your Priorities for Java and the JVM?
InfoQ's research widget has been deprecated. It should continue to work however, and we hope to relaunch it at some point in the future.
With Oracle investing heavily in Java, and a roadmap for Java 9, 10 and beyond under discussion, we thought it would be interesting to get your views on what the priorities should be for Java the language and the JVM as a platform. InfoQ is experimenting with a new tool for surveying the views and opinions of our readers on important topics.
There are 17 items on the list, and for more background on many of these ideas see:
I would have loved to vote ...
Re: I would have loved to vote ...
The voting widget has been developed as a completely separate service, eg similar to Disqus, so that in the future we'll be able to embed it in any page we like without having any dependencies with the InfoQ CMS. For example you will be able to embed it directly in your own blog.
We're also trying to see what would be the best way to handle authentication, for users that already have an InfoQ account.
For the record, if for example you chose to vote with your FB account, the service would only be getting your name and email address, and not information like contacts, nor would it require privileges for posting stuff on your timeline.
Thanks for taking the time to send us feedback!
Asking for way too many privileges here
Re: Asking for way too many privileges here
I'm replying inline:
Why does the survey want us to login using another authority?
As I mentioned earlier in this thread, the voting widget has been developed as a completely separate service (different domain), eg similar to Disqus, so that in the future we'll be able to embed it in any page we like without having any dependencies with the InfoQ CMS. For example you will be able to embed it directly in your own site, blog, etc.
And why does it want to manage my contacts on Google? Or see who is following me on twitter?
Very good observation about Google! I've just shipped an update that requires only the absolute minimum permissions for Google. As far as I understand, the current Twitter permissions are already the bare minimum.
Thanks Patrick for taking the time to send this feedback. It's very valuable for us in order to improve this service :)
All for JVM, nothing for Java is what I want
Improve the JVM and only patch Java, no more wasted effort should be expanded on it
Won't vote with this authentication scheme ... / question unclear
Nasty, that I therefore cannot see the analysis nevertheless >:-(
Beside: The "mobile support" question is not intuitive. What exactly will I tell you if I vote that item?
Will I vote for another mobile effort, or will I vote for staying with ME?
Example: JS, Obj-C/C++, C#,... but no JVM on iOS!
Server-side UI frameworks are going to die.
I don't understand why client-side development is so ignored.
I wonder how any of those new super-cool JVM languages could be useful without a JVM to run it on major mobile and desktop OSes.
As for Segmented stacks (and coroutines), it seems they are a good way to write very simple asynchronous code,
instead of the complexity of Promises, callbacks spagetti, etc.
Probably any JVM language could implement it at the compiler level like C# does (async) but it seems inefficient (like emulating closures with inner classes).
Re: Won't vote with this authentication scheme ... / question unclear
"Will I vote for another mobile effort, or will I vote for staying with ME?"
For the purposes of this survey I wasn't really looking to drill into the specifics of whether any mobile effort should look at creating a successor to Java ME or continue under the ME brand, but whether InfoQ readers felt it was worthwhile Oracle investing in the mobile/embedded space at all or whether they should leave it to Apple and Google, and instead focus on the core Java SE/EE technologies and underlying platform.
What about hot deployment?
Too low level!
Re: Too low level!
If Scala already have the features (from the votes) why on earth do we need Lambda in Java? Why not making the opportunity to improve Scala instead instead of taking more years to release? It doesn't make sense to reinvent Java syntax whereas C# is for OO and F# is for Functional.
1) An author of the list has exactly! the same priorities as average voter has ;)
2) The results are strongly influenced by ordering of the options (voters simply go from the beginning and have less and less points to use and don't remove earlier placed points frequently).
3) The ordering is changing in time, by using already collected results.
4) Something else I can't see.
I can't imagine 1). 2) would mean wasting of time of 700+ voters and 3) would be a joke :) so I hope 4) is right.
The regular set of unsigned "machine types"
Value classes including arbitrary precision integer (that starts as a "small-integer" tagged into an oop) and C-like packed structures with alignment that can overlay byte arrays.
Introduce/deprecate covariant arrays
Remove null references
IEEE 754-2008 | ISO/IEC TR 18037 | Floating Point | Fix Point | Q / Rational numbers