The dawn of enterprise mashups has been breaking for some time. It's an interesting discipline that has proceeded in starts and stops since its initial hype phase. Consumer applications built with mashup techniques have become commonplace and yet enterprise mashups are still hovering somewhere just outside the mainstream.
Many mashup discussions still start by defining what enterprise mashups are, which might be a necessary evil considering how easy it is to mix up enterprise mashups with music mashups and others.
Consumer mashups are cool and edgy; they're quick and easy to make; they often make you say "wow!"; and sometimes they can save you time on researching where to get lunch or even help narrow your housing searches; but they don't normally serve functions that are of express value to the enterprise. Unfortunately, this may leave many architects with the feeling that enterprise mashups are also cool and edgy; quick and easy; and ultimately not very valuable.
It doesn't help that enterprise mashup standards groups like the Open Mashup Alliance (OMA) have low visibility and are slow moving. However, the OMA's Enterprise Mashup Markup Language (EMML) is probably the most available readymade toolset for developers who are looking to start into enterprise mashups. But enterprise mashup developers may not be interested in readymade tools.
Enterprise mashups are kind of an art form, just like hip-hop or techno mashups. The developers who produce them are creative artists with the technical savvy to understand the inner workings of multiple enterprise applications (and in some cases outside applications and other resources as well), the vision to see how they could work better together, and the patience to make it happen.
There's no guarantee that any two applications will benefit from being mashed together. In some cases, an enterprise mashup can be much more than the sum of its parts. In others, the individual applications work better separately than they ever could together. Enterprise mashups have to be taken on a completely case-by-case basis, and – without that very particular skill set – each case can be difficult to evaluate, which is likely why the business side is hesitant to buy into enterprise mashups and why the discussion on enterprise mashups hasn't really moved forward in recent years.
This was first published in January 2011