What single word do you think of when you think of Web 2.0? (If you think about it at all.) I think of the word "social" but maybe that's because I'm a social user of Web 2.0 applications, and not a technical developer.
So, when I wrote a few months ago about
the future of BI and Web 2.0, my two envisioned categories in this area were both socially oriented:
- The sharing of BI output via communities and other 2.0-type sites.
- The analyzing and reporting of textual content and behavioral patterns on social media sites.
When
Computerworld talks about the future of BI and Web 2.0, however, they get technical.
They talk about mashups:Welcome to Business Intelligence 2.0, a world in which one of BI's original big promises is finally being met, and a broader class of everyday business users — as opposed to statisticians or data analysts — are tapping into innovative technologies and Web-based BI capabilities. Police officers, physicians, accountants and salespeople are mashing up and analyzing structured and unstructured data from far-flung sources in the ways that make the most contextual sense to them.
Here's where I admit I'm still working to understand mashups. What do I know? I know mashups have something to do with combining two or more applications or Web sources to create a brand new application. I'm pretty sure I'm using mashups when I play with
fd's flickr toys,
twistori and similar apps - but these are not business related applications. I know a lot of people are using
Google maps and other publicly-available mapping applications to create mashups. That's where I hear the term a lot: in conversations about Google maps (and, of course, in conversations about the book
Wikinomics).
Likewise, a lot of the examples in the
Computerworld article talk about geographical data sources. In fact, when I first read the
Computerworld piece, I was reminded of the article
Predicting the where with Spatial Intelligence from the fourth quarter 2007 issue of sascom. So I contacted the author of that article, William Holland, to get his thoughts on mashups.
William is the CEO of
GeoAnalytics Inc. and an expert in implementing geographical information systems (GIS). I started by asking if any of the customers mentioned in his article were using mashups. He responded via email and wrote, "Technically those implementations are not 'mashups' as we would define that term in the context of spatial (geographic) information systems."
In his world, William says, a mashup is a "combination of a set of web services, especially map services, that are drawn from multiple sources on the fly." For example, he says, you could use Microsoft's
Virtual Earth with
ESRI ArcWeb Services for demographic and thematic map data - and combine both with internal customer data, including locations and other characteristics, for analysis.
However, William says the examples in his sascom article aren't mashups. Why? Because even though the applications are consuming internal and external Web services, they ultimately combine and store the data in the customer's own repositories.
William goes on to say, "There are many reasons why clients take that approach, but primarily it is for security and data quality reasons. For these clients, that is because both spatial and non-spatial data needs to parsed, cleaned up, and harmonized."
What he's saying here mirrors the concerns of Mark Hall, who writes,"
Approach mashups with caution" in the same issue of
Computerworld. When considering mashups, he asks:
... do you know if the originating systems meet your security standards? Are those systems at current patch levels? If your business works in a regulated environment, will such a mashup put you out of compliance?
It's worth thinking about.
I'm left wondering what SAS customers think of mashups. Have you built a mashup application using SAS? Do you have an idea for a fun SAS mashup? Would your IT department allow it? Or are the security concerns too large at your organization?
