On this podcast Charles Humble and Wes Reisz talk about Java 9 and beyond, Kotlin, .NET Core 2, the surge in interest in organizational culture, quantum computing, and more.
Key Takeaways
- Java had a big year with Java 9 shipping, Java EE going open-source and moving to Eclipse as EE4J, and IBM open-sourcing J9. From next year the platform will also be on a bi-annual release cycle with the next two versions (expected to be Java 10 and 11) both shipping during 2018.
- Kotlin joined Scala, Clojure, and Groovy as a strong alternative language for the JVM, particularly for mobile where it was buoyed by Google’s official blessing of it as a language for Android development at Google IO.
- InfoQ saw a big surge in interest around .NET linked to .NET Core 2.
- At both InfoQ and at QCon San Fransisco we also saw an upsurge in interest around organizational culture, with one of the culture tracks (the Whole Engineer) moving to one of the larger rooms.
- We started to see Quantum computers emerging from the labs, with IBM making a 16 Qbit quantum processor available via their cloud for developers to play with, and the corresponding library available for Python on Github,
- Another major trend from the year was the availability of machine learning libraries for software developers to build and train models.
Subscribe on:
You are the chief editor of InfoQ. What does that entail?
- 01:20 It’s broadly two different things; making sure that we’re producing content across the various different formats (videos, news articles, podcasts and so on).
- 1:40 I work with the marketing team on any promotions that we’re doing, and work with the freelance team which is the 50 or so writers that produce content for InfoQ.
- 1:50 The other part of the job is more strategic; thinking about what topics we should be covering and what’s coming.
- 2:05 It’s a bit like being an analyst; it’s looking further ahead to what’s coming, and finding out of the things that are being talked about and hyped up at the moment which ones are likely to stay.
InfoQ is really well known for architecture, Java, .Net and culture coverage. What has happened recently in these spaces?
- 5:20 Java 9 shipped this year with the Java Module framework (formerly known as Jigsaw).
- 5:35 The other key change that happened was Java 9 shipped with the G1 garbage collector by default.
What about Java 10?
- 5:55 Java 10 will be out in a few months, and move towards a bi-annual release cycle.
- 6:10 Java 10 will be out in March followed by Java 11 later in the year.
- 6:30 If you’re a large enterprise, and you’re struggling to keep up with a major release every couple of years, I wonder whether they will be able to keep up with a change every six months.
- 6:35 Obviously the releases will be smaller, so if you look at what’s coming in Java 10, local variable type inference is one of them.
- 6:35 There’s also some interesting changes coming with the JIT compiler; Graal is an experimental JIT for Linux.
- 7:05 It’s not a huge release, but it’s hard to know how it’s going to play out for the big enterprises - it’s hard to know at this point.
- 7:40 Historically there’s been a huge problem with the big monolithic releases that have kept slipping.
- 7:55 If they can bring down the size of changes then it might work.
- 8:00 If you’ve got to test a number of large enterprise applications, then that’s a painful process.
- 8:15 I admire Oracle for trying it but I don’t know how it’s going to work out.
- 8:20 Oracle’s stewardship of Java has been a lot better than people anticipated - there have been bumps along the way - but generally they’ve done a good job.
- 8:40 I don’t know how well it will work in practice, because we haven’t tried it yet, but two big releases coming up this year will be interesting.
- 8:50 If it allows Java to evolve and move a bit quicker then on balance that’s probably a good thing.
- 8:55 A couple of other things that have happened - IBM opened up the source for their Java runtime (OpenJ9), and JavaEE got open-sourced and moved over to the Eclipse foundation.
- 9:05 JavaEE has fallen a bit behind the needs of modern developers, although it doesn’t mean it doesn’t have its uses.
- 9:15 It wasn’t designed for a cloud/micro-services environment, and increasingly that’s how software is being built.
- 9:20 Moving JavaEE to open source and to the Eclipse foundation is interesting.
- 9:25 Microprofile was initiated by RedHat to do micro-services for Java has also moved to the Eclipse foundation as well.
- 9:40 Maybe JavaEE (or now, EE4J) and microprofile will merge in some way in the future.
- 9:50 It gives SpringBoot and the CloudFoundry people something to compete with, which is not a bad thing.
- 9:55 SpringBoot and Spring/Pivotal in general have been doing very well in the meantime; SpringOne was in San Francisco’s Moscone West, which was the biggest they have ever had.
- 10:15 It’s been a good year for Java.
Kotlin has been doing well this year as well.
- 10:40 Kotlin got blessed by Google as an officially recognised language for Android, and that’s helped the language gain momentum.
- 10:55 The fact that it looks quite similar to Swift probably doesn’t do it any harm.
- 11:00 If you’re developing for iOS in Swift and are developing in Kotlin for Android, they look quite similar, and that doesn’t do them any harm.
Kotlin native is doing well as well.
- 11:30 It’s interesting to have a new language on the JVM that has caught on in the same way that Groovy or Clojure have done.
- 11:40 It seems to have found a definite niche of developers, and having JetBrains behind it doesn’t do any harm.
- 12:05 In some ways the way that Microsoft was able to integrate Visual Studio and .Net is similar to JetBrains’ IDE and language.
You mentioned .Net - what’s happening with that?
- 12:30 It’s seen a huge uptick this year.
- 12:35 Microsoft as a company seems to have really changed - the change of leadership there has resulted in a significant change in attitude.
- 12:45 .Net core really seems to have caught people’s imagination - being able to run a subset of .Net on Linux is really appealing.
- 13:05 C# is a really nice language - similar in a lot of ways to Java.
- 13:10 If you think of InfoQ’s audience as being more open-source oriented, for a long time .Net wasn’t an option - and suddenly now it is.
- 13:30 The .Net Core 2 that we wrote about over summer caused a surge in interest.
- 13:35 I think it’s a really good time for .Net and Microsoft in general.
What’s happening in culture at InfoQ?
- 14:15 We’ve been writing about culture and agile that something InfoQ started with.
- 14:25 It’s an important part of our content, but there seems to have been a surge of interest.
- 14:30 It may be that the technology industry is waking up to it.
- 14:35 You also can’t ignore the fact that in 2017 there have been a few uncomfortable difficult stories around ethics in software.
- 14:55 There’s been a huge diversity story running this year; not exclusively in the technology sector.
- 15:05 We all know that sexism in the industry is a problem - the fact that we’re talking about it is a good thing.
- 15:15 We know that diversity matters and will produce better products.
- 15:25 It’s important that InfoQ have a culture podcast as well as this one, and that’s been adding listeners.
- 15:35 We saw in QCon San Francisco that the culture tracks were doing really well.
- 16:40 Finding people, retaining people and motivating people in an area where recruitment is so strong is a key competitive advantage.
- 16:50 It’s not just having really talented people; it’s having an atmosphere in which they can feel comfortable in and able to produce their best work.
What big stories from 2017 surprised you?
- 17:10 We ran four or five stories on quantum computing, which I didn’t see coming.
- 17:15 If you’d have asked me the question last year, I wouldn’t have quantum computing on my list.
What is quantum computing?
- 17:35 It’s really difficult to describe.
- 17:40 Whereas a conventional computer has a bit which is 0 or 1, a quantum computer can represent both simultaneously.
- 17:55 In theory, it allows you to do certain types of computational task much more quickly than are currently possible with conventional computers.
- 18:10 The area that gets the most focus is cryptography; most encryption works on asymmetric keys where if you have a fast enough computer then all of that will cease to work.
- 18:35 The thing that surprised me wasn’t that Microsoft, IBM, Google, DWave were building these systems, it was the fact that you can go to GitHub and download a Python library, sign up to IBM’s bluemix and play with a 16 qubit processor for free.
- 19:05 It’s not a hypothetical thing that doesn’t exist - it’s something that exists now.
- 19:15 The quantum supremacy stuff is some way away - we aren’t at 50 qubit systems yet - but people have started to build these things and have access to them today.
What about AI?
- 21:10 This year has really taken off - it feels like this is the year that AI and ML have exploded onto the scene.
- 21:20 What’s happened this year is that a lot more accessible ways of exploring this space have happened.
- 21:35 It also feels like there are a bunch of problems that we know we can solve to some extent with this technology.
- 22:00 If I can get a computer to categorise photos or recognise people then that’s a big deal.
- 22:10 There are whole categories of problems that have been unsolvable - like being able to search the video archive - but if you could understand the text with speech-to-text then it becomes much more powerful.
- 22:30 We’re launching QCon AI next year to bring some of the people together.
What’s your technology focus for 2018?
- 23:00 The big thing that I want to do is to try and find a way of making the InfoQ video archive searchable.
- 23:00 My idea is to feed the audio through an NLP engine and search through for Lucene, for example.
- 23:45 I also want to investigate more in Java 10 from a developer perspective.
You mentioned at QCon San Francisco that you wanted others to share and talk. What did you mean by that?
- 26:00 I feel like it’s an exciting time in software - we have extraordinary amounts of software and incredible amounts of computer power at our fingertips.
- 26:25 There are amazing things that we can do - it’s one of the most interesting times in software that I can remember.
- 26:40 There are also some risks that come across with that - with great power comes great responsibility.
- 26:55 When I started in software there wasn’t really the concept of open source, and people didn’t really talk to each other.
- 27:00 Even within companies it was hard for developers inside a company to talk to each other.
- 27:10 I think that’s changed to a great extent.
- 27:15 I think our industry is absolutely at its best when we are sharing between each other.
- 27:30 It doesn’t have to be to speak at QCon - write a blog, present at a meetup - just share your knowledge and experience.
- 27:40 As an industry it’s how we make progress and improve - and also how we can avoid some of the mistakes.
- 27:45 It’s something I believe in passionately - we were talking about new year’s resolutions, so I have made one to put together a soft skills talk about managing distributed teams, which is what I do now.
- 28:10 That’s what I would urge everyone to do - think about what you have learnt and how you can share it with the wider world.