Blog post

Docker is the New Java

By David M. Smith | October 21, 2014 | 1 Comment

Microsoft and Docker’s recent partnership reminded me of how the world has changed. I’m thinking about the introduction of Java back in the 1990s. Back then I was extremely skeptical of Microsoft’s motivations for their “support” of Java. This time I am not skeptical of their motivations with Docker.

The more I think about it the more similarities I see between what was going on then and now, it seems to me that Docker is the New Java (an analogy – yes I know they are very different technologies – I’m comparing the similarities and the roles they are playing in the industry)

– Both came on the scene extremely quickly, going from obscure to deafening hype in 12-18 months
– the main value proposition and hype for both is portability (and they will both not deliver on 100% portability)
– the first excitement was in the development community
– in their early days, people are trying to find production implementations, not finding them, and saying that the technology is not ready or just for internet or consumer use – not for real enterprise use
– both were touting “open” (even though originally Java was not open source)
– both were talking about containers – in the case of Java it was the Java virtual machine and bytecode
– vendor dynamics were playing a huge role in the hype and the alignments
– vendors walk a fine line in supporting an open technology with the value add and differentiation they need
– There are higher-level areas where disagreements, complexity, and nuance will thrive. In the case of Docker, that level is orchestration (e.g., Kubernetes). In the case of Java there were attempts to position it as “just a language” when it was much more.

Some things never change though. Vendors support things because it is in their interest, not those of developers and users. If those interests coincide, fine. But that’s not what drives them. Certainly there is usefulness to developers and that is driving excitement. But as for broad support by vendors, it is driven not just by usefulness to developers but usefulness to vendors. Which means competition, FUD, etc.

The potential to vendors of Docker and other OSS and portability enablers (e.g., CloudFoundry and OpenStack to some extent) is to change the competitive dynamics of the market. Those leading markets (like Microsoft did in the 1990s) don’t want to see change while those challenging (Sun in the 1990s) look to leverage the new into the change they want.

The announcement from Docker and Microsoft highlights how the market has changed. Back when Microsoft ‘supported’ Java shortly after its announcement, they were the market leader and they fooled lots of people. Today Microsoft really does support Docker. Because they are (along with Google and others) trying to wrest workloads, leadership, etc. away from Amazon (the public cloud leader) and WMware (the private whatever leader).

When I hear that VMware and Amazon also ‘support’ Docker, I need convincing. I also think, like Java, Docker will play out over several years. Many are overerestimating its impact today, but longterm likely to be underestimating it.

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed

1 Comment

  • Clearly Docker is the current industry “fashion”. From one perspective there is little that underpins it — over and above making Linux Containers more friendly to use. The issues of image sprawl and artefact management are not – IMO — appreciated by the industry and many of the same mistakes made with Virtual Machines – will be – yet again replicated. However Docker breaks the strangle hold in the Market created by Virtual Machines and we should be thankful for this.

    The value of increased artefact modularity in the runtime is slowly being understood. This direction is fundamentally correct – and at some point
    OSGi will be the new hype – as it addresses many of Dockers weaknesses – I look forward to that day.