Editorials

Single Source of Truth – A Challenge for SQL Server

Featured Article(s)
SQL Server 2008 Unstructured Data and Filestream
SQL Server 2008 Unstructured Data and Filestream

Single Source of Truth – A Challenge for SQL Server
For quite some time now I’ve been writing about, and Microsoft has been touting the concept of a single source of truth (phrased different ways). The concept has to do with knowing your information is "valid" from a known-good, single source. Instead of just taking those sales numbers at face value, it’s caring about the source of the numbers, making sure they come from good sales data, for example.

Single-root-sourcing information is important because you can get to the point where you don’t know *what* to trust or believe because information has been touched by so many people. It’s a bit like the old "too many cooks in the kitchen." All of this tripped in my head as I was reading a review of a new search tool called WolframAlpha. (Read the review here) Funky name aside, the reviewer was talking about a hesitation to rely solely on the tool to find the best single source of information because they (the reviewer) didn’t trust information that couldn’t be confirmed across multiple sources.

I get that – but it made me wonder… how does this thought process impact how we look to single-source information in our systems? Perhaps it’s a stretch to make the comparison, but it goes a bit to the thought process many people are growing (or have already grown) accustomed to. That process is the Google thing – that of having lots of results to compare/contrast when considering a specific issue.

I’m not saying we need to use Google to source our information, but has mentality changed to where people will actually not *trust* a single-sourced information bit? If the general public (and hence your users) approach has changed to getting information, will this impact our application development? If so, how? If not, why?

What do you think? Drop me a note – does this impact things in the near future? Do we need to consider changing how we go about serving up information? Does this apply only to "research" and not necessarily to transactional data and reporting?

Email me here

Featured White Paper(s)
TECHNIQUES FOR IDENTIFYING AND OPTIMIZING RESOURCE-INTENSIVE SQL SERVER QUERIES
Analysts are often provided with vague descriptions of database system performance like its sluggish or this report takes for… (read more)