Thoughts for the Weekend and TechEd
I wanted to return to my thoughts earlier this week about questions, our data and what types of things we have probably in the future in terms of needs for the data.
That’s a mouthful, I know.
As we head into TechEd, see all the new gadgets, hear about the new versions of software and features, I think we need to start changing what we are looking for and expecting from our information systems.
I think we have it pretty solid when it comes to storing information, pulling it back, tweaking it, etc. But I think the value in our information systems comes from our learning from the information on file. You’ve heard of decision support systems, and I’ve even heard of (though am ignorant about) inference engines… but I think we need to start paying attention to the next era of information and data. We can relate columns, tables and information bits until we’re blue in the face – very cool!
But we need to start having the systems be more helpful. We need help determining the questions to ask of our systems. One example is our own reporting for our work with SSWUG, our conferences and such. We were looking to better understand the types of things that people are looking for in the content. Figuring out the questions to ask is extraordinarily difficult. Why can’t our database servers move on to that next level? Why can’t they recognize, for example, that several tables have member IDs – and that we can figure out all sorts of better ways to present information and content based on how people request things from the site.
It’s one, small, silly example. Think what you could learn! But that’s the point. We don’t know what we could learn; we don’t know the questions we could ask, answers we could receive.
I hope that SQL Server version xxx will start to feature options that help us use information, not just store/retrieve. We need the functionality we have today, no doubt at all. We need the new things coming down the pike – no question. But what if you were able to get a list of analysis queries that can be answered, without you having to create the questions.
Select * from vw_learnings
Learnings
———————————-
Customers with orders by month
Customers with orders for linen supplies in North America by date range
Employee vacation time usage relative to weekends, holidays
Days of week most popular for coffee flavors
….
99 row(s) affected.
You get the idea – look at the data – what relationships are there and how could we use them? Help me *use* the data.
Webcast: Best practices for Analysis Services administration
This session takes you on a end-to-end tour of management features for Analysis Services. Along the way, you’ll learn important best practices direct from the product team about how to configure, optimize and manage your Analysis Server. We expect this to be an essential session for new and experienced Analysis Services administrators alike.
Presented by: Donald Farmer
> Register Now
> Live date: 6/23/2010 at 12:00 Pacific