Faster Java: Do We Really Want It?
It would seem intuitive that consumers of technology like to have frequent releases that introduce additional features and fixes, right? Maybe. But maybe not.
Join the DZone community and get the full member experience.Join For Free
Last week I was at the Voxxed Days Bristol event, which was as enjoyable as last year. Bristol appears to have a lot of developers, and there is clearly a thriving startup community there. As ever, there is plenty of interest in Java.
Originally, I had been scheduled to deliver my talk on “55 new Features in JDK 9”, something that has proved to be pretty popular since I put the slides up on Slideshare. However, when I arrived in Bristol the night before the conference, I got a message from the organizers asking me whether I could deliver the keynote presentation. Unfortunately, due to unforeseen circumstances, the originally scheduled presenter was no longer available.
Not being one to pass on an opportunity like this I happily agreed. The suggestion from the organizers was to use my intended session, but I had some ideas and suggested I talk about “Is An Agile Standard For Java Possible?” (Slides are also available on Slideshare)
The theme for this presentation was the same as a blog post I’d written earlier this year. In my presentation, I discussed the ways in which the standards for the core Java platform have been developed in the past (primarily through the JCP) and how this might change in the future.
What was most interesting, and revealing, to me was when I did a quick poll of the audience. Assuming we were to switch to an agile approach for developing and releasing new features for Java how frequently would people want this? I started with the idea of a quarterly release, which got less than a handful of votes. The majority of the audience seemed happy with the current rate of release. Obviously, this is only a small sample size, but I suspect it’s fairly representative of developers in general.
I discussed this point with some people afterwards, and the general opinion was that sure it would be good to have access to new features more quickly, but the downside is the amount of testing, certification and support implications that agile development of the platform generates.
During my travels, I will certainly be gathering more data on this question because it may well be that an agile approach is appropriate for applications, but not so desirable for the core Java platform. Let me know whether you feel strongly one way or the other.
Published at DZone with permission of Simon Ritter, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.