and now for something completely different...
In early 1998 I was working as Dev Manager at a startup in Scotland building tools for Data warehouse Design and Management, for the most part we targeted Oracle (RDBMS and Express, which they had just acquired from IRI) and Business Objects, with a little work on Ingress. We had one customer running SQL Server 6.5 but it was not a big system and they were always an afterthought with new versions as our focus was on the "real" db, Oracle. We had heard about "Sphinx", "da Vinci", "Plato" and "Tensor" but until Microsoft UK invited us and a bunch of other Oracle ISVs to a preview of Beta 2(in of all places Falkirk Police College!) we had no interest. To say the impact of seeing these technologies was huge on us(as a company and for me personally) is an understatement, it was clear that Microsoft had the intent to execute on an industry changing set of strategies around BI for the Masses and TCO/Ease of use.
We came away unsure of what to do, we had been working on the assumption that we could continue to live in the boutique/high cost BI world that might not now exist. We continued on plan but a couple of us started a skunkworks to play with the bits and see if the reality matched the promise and what to do about it. By the summer we had a prototype of our app running against "Sphinx", there was no/limited documentation on the DTS, DMO, ADO MD or DSO APIs, we worked by using trial and error against the COM interfaces, but we ended up being able to do FAR more against the beta of "Sphinx" than against RTM Oracle as the interfaces were all exposed and much richer. Our prototype allowed us to build Databases, DTS packages, OLAP Cubes and Excel Pivot tables all programmatically from our app, derived from our dimensional model. During this time it became clear that SQL Server 7.0 and beyond was going to fulfill its promise(and more)and I felt that I could contribute to that promise, hence I found myself in Redmond wandering back and forth between building 1 and 3 for interviews, and the rest as they say is history.
Looking back, the "Sphinx" platform achieved everything it set out to do and more, I believe the rationalization of the BI platform industry is a direct reaction to Microsoft's ambitions in the space, I believe the focus from the other major vendors on ease of use and manageability is a direct result of Microsoft's demonstrated leadership in game changing use of self tuning algorithms, automation and GUI. The recognition of SQL Server's success comes in many dimensions, the TPC-C wars of the early 2000's, the implementation of certain features and certain names/limitations of features/products are a form of flattery from competitors (Thanks Larry…) but the true recognition comes in the form of the customers who are able to run their business more effectively, by using the services provided by SQL Server. The sheer volume of apps, customers and data is far grander than any thought possible 5-10 years ago.
As we look forward at the potential for SQL Server it is truly amazing(with the last 2 quarters of > 30% growth being one small indicator…), however for me it seems the future lies elsewhere, after 10 years of working on database systems and tools, today was my last day in the SQL Server Team. On Monday I'll be starting a new position in Visual Studio working on Team Test. As such this blog is going to get a little quiet while I get up to speed on my new role, I still have lots of SQL Server content left to post which I'll get to eventually, but it will not be as frequent as it has been.