OData and Microsoft BI
I first came across OData last year when Jamie Thomson blogged about it; since then I hadn’t really thought about it much until I came across Douglas Purdy’s blog and specifically his post from yesterday which really brought home how important it is to the future of Microsoft BI. I would urge you to watch the video “OData: The Movie” that he mentions in his post because it gives a really good introduction to the way that OData can be used in BI scenarios; if you don’t have the time, all you really need to know is that it’s OData that makes it possible for PowerPivot to consume data from SSRS reports and Sharepoint lists.
Just watching this video made my mind boggle with the possibilities of OData, although since I’m a long way from being an expert in this area I won’t bother to detail all of these fantasies as they’re probably rubbish. However, just consider how much easier life would be for the PowerPivot user of the future if the internet was full of sites that supported OData; certainly, when I’ve looked at the new UK government websites such as the recently-launched http://data.gov.uk/ and http://data.london.gov.uk/, or even sites like the Guardian Data Store that I blogged about last year, I’ve felt that the lack of a standard format to consume this data (a badly-formatted Google spreadsheet is not ideal) has seriously limited these sites’ usefulness. And what if Bing came up with a service like the sadly-useless Google Squared, where you could search for data on the web and return it in a structured, OData format?
I’d also love to see Analysis Services support OData too, in some shape or form. Perhaps it could be used to solve some of the same problems with XMLA that Julian Hyde lists in his recent post on xmla4js. Wouldn’t it also be cool if there was a standard interface for publishing an MDX query up to SSAS from any client tool, and then be able to consume it via OData (similar to what I suggested here)? You’d then enable scenarios like this: power user creates query using an advanced SSAS client tool, then publishes it up to SSAS whereupon it becomes something similar to a SQL stored proc with the results available as an OData feed, and therefore can be consumed either in your own code (eg in a website), by PowerPivot users, or by something like SSRS (which would then know nothing about the query used but would just be used to format the resultset).