This weekend, I listened to a TechTarget podcast interview with Michael Ogrinz, author of "Mashup Patterns: Designs and Examples for the Modern Enterprise." The title of the podcast, "Enterprise Mashup Patterns Can Make Data Integration Easier," naturally had caught my attention, but what actually stood out the most to me during the discussion wasn't about integration specifically.
Instead, I was taken by Ogrinz's own experience with mashups.
Ogrinz is a an enterprise architect, specifically a principal architect for global markets at one of the world's largest financial institutions. He encountered a company demonstrating its mashup tool at a conference, but the presentation focused largely on its use with Web data and tools. After seeing the sales pitch, he told the reps they should give his company the $100,000 tool for free because it would never be able to use the tool.
Ah, but the universe has a way of making us eat our own words, doesn't it?
When he returned to work, he kept thinking about mashups. Soon, he realized he could use mashups inside the enterprise to address many of those persistent, unaddressed problems that, viewed as conventional IT projects, were too big, too expensive or too darn technically hard to justify:
All of a sudden when I realized there was a new set of tools that could address these unmet problems, then I started seeing potential for mashups everywhere. All these things over the years that I thought there was no way to extend an existing system without the original developers involved, or there's no way to add an RSS feed to an system unless you have the source code for it, there's no way to change its interface, there's no way to fix a bug unless you have access to the original developer.
This is what caught my attention. After all, I'd explored before about <strong>how mashups could be used for simple data integration</strong>.I'd even read a previous argument by Ogrinz that mashups would replace portals-a position he actually backs down from in this interview. But much of what I had read focused more on mashups as a way to get those pesky little projects done that otherwise weren't worth doing.
This is the first time I've seen something that really positions mashups as a means of addressing enterprise IT problems that had, using conventional approaches, simply been too hard or too expensive to tackle.
The good news, though, is that the tools vary so much, there's probably a tool that can fit your organization's needs and budget, he says.
Data mining remains a low-hanging fruit for using mashups-a way to realize a quick ROI and justify the expense, according to Ogrinz, whose book offers a lot of data-mining mashup patterns. As he explains it, he's not offering technical patterns, but rather 34 business-focused, non-technical patterns where mashups can provide tools.
The interview includes a link to a sample chapter of Ogrinz's book. Unfortunately, it's a bit of a challenge to read due to some weird formatting issues, which I hope are caused by some issue with my version of Adobe Reader. It's 51 pages in all, but that's with huge spaces and gaps. For my money, the best stuff started on page 13.