A killer app is any product that is so desirable that it fuels the sale/adoption of a larger technology that the product depends on.
The gaming industry has been aware of the power of killer apps for a while, and often seeks to create launch titles for their new gaming systems that are killer apps. Microsoft, for example, deliberately created the first Halo game at same time as their Xbox console. And Nintendo has repeatedly introduced new Mario titles with its new consoles.
However what may not be so obvious is that killer apps exist for learning new software platforms as well, such as for APIs, SDKs, and programming languages.
Learning a new platform (along with the associated tools and ecosystem) takes a large investment of time, and so there is a similar barrier to adoption as, say, the monetary cost of buying a new game console.
I think the platform-builders in the software industry could learn from the gaming industry and strategically introduce killer apps alongside new platform offerings. I can think of no prior case where this was done deliberately (see the Appendix below), so I think our industry (i.e. software) is really missing out.
Killer apps for languages:
Killer ecosystems for languages:
Killer features for languages:
More information about the adoption of Lua in the gaming industry on StackOverflow: Why is Lua considered a game language?↩
Although Microsoft has finally created a version of Outlook for the Mac, it is buggy and missing a number of advanced features of the Windows version. Thus two of my colleagues who do a lot of planning run Outlook specifically on Windows despite the workplace being otherwise Mac-oriented.↩