For the wrap up of the series of SQL Server 2005 glorifications, I am writing about couple of things I do not like too much in the product. This way I hope the series is going to bee slightly more balanced.
I am sorry Microsoft did not decide to make the UDM Perspective a securable. It is very nice to have more than one fact table supported inside a UDM cube, but it means also more complicated security. Perspectives could simplify administration a lot. Unfortunately, you cannot set permissions on perspectives. Therefore, I expect a single UDM cube is still going to cover a single business area. Still, having multiple fact tables inside a cube is good, because you can have actual and planned data together.
The SSIS expression language is confusing me. Similarly, SSIS data types do not impress me. SQL Server 2005 in its entirety already supports many languages: T-SQL, CLR languages, MDX, DMX, XQuery, XMLA, scripting languages… We can say the same for data types: SQL Server types, CLR types, XML types, Analysis Services types (double types – for OLAP and Data Mining!)… So why do we need another language and another set of data types? I would prefer if SSIS expressions would be either T-SQL or VB.NET based (here I would actually really prefer VB.NET to C#, because scripting task and transformation in SSIS already use it), and if data type system would be either SQL Server or CLR based.
From the tools point of view, I simply miss the good old Query Analyzer. The SQL Server Management Studio (SSMS) and the Business Intelligence Development Studio (BIDS) are great tools, but because they both base on Visual Studio, I feel them as exaggerated many times. They are too slow. Besides that, I am not creating a solution always; sometimes, I just want to create a problem. In addition, some overlapping tools make confusion. For example, when do you use Surface Area Configuration tool and when Configuration Manager?
Pivot operator is not dynamic. Pivot operator is not dynamic. I am repeating this intentionally, not because I would be from Baden-Baden.
Finally, Books OnLine is far from the quality of SQL Server 2000 Books OnLine. I spend much more time for searching now then I used to in the past.
- Python for SQL Server Specialists Part 4: Python and SQL Server - April 24, 2018
- Python for SQL Server Specialists Part 3: Graphs and Machine Learning - April 11, 2018
- Python for SQL Server Specialists Part 2: Working with Data - March 22, 2018