I've historically been pretty bullish on the whole concept of mashups, particularly end-user mashup technologies which promise to offer the next step in offering the average user programmer-like power to process and report generation. If this sounds outlandish, as many claims do which promise to offer programming without coding, consider that some of the most popular office applications have done exactly that at other levels of abstraction... spreadsheets are just a way to program complex mathematical formulas without coding, and word processors are just a way to do layout and formatting without hand-coding it all for a particular machine and printer.

But I haven't really seen the promise being fulfilled yet in the enterprise, and so Dion Hinchcliffe's recent post covering the state of the mashup industry and predicting its massive expansion, while an excellent resource, leaves me a bit clammy. Clearly if all these serious movers are getting into the market other people think there is something to it as well, but as Hinchcliffe points out there remains a significant gap between the expectations and capabilities of users and those of the mashup engines. Maybe this is just the same sort of gap that we saw in many offices when PCs, with their personalized productivity applications, were deployed, and so maybe it's just a generational thing that will cease to be an obstacle when more mashup friendly kids get into business. On the other hand, while there are a lot of cool Web 2.0 mashup tools and examples out there, most of them are really just that: cool, not tremendously useful. Is the utility coming to enterprise mashups? Or is it just a trendy application of Web 2.0 for the enterprise that's more sales gimmick than practical tool?