Hi Folks, I’d taken a bit of a break from producing SQL Down Under podcasts over the last few weeks. It came on the end of a lot of travelling. However, there is good news for listeners. I’m recording three shows over the next week, so watch out for a bunch of new shows. I hope you really like them. Another question I’ve had from a few people is why I don’t get someone to interview me for a show. I remember Kent Tegels offering to do that a little while back and I might take him up on it soon. … Continue reading SQL Snapshots – coming soon
In answering a query on the SQLDownUnder mailing list today, one of the functions Paul Nielsen mentioned was the NewSequentialID() function, which is new to SQL Server 2005. I then had an email asking about this function, so I thought a short post explaining it might help. In a previous post, I discussed one of the problems identified with using GUIDs as primary keys where they end up being used in clustered indexes. The preference for clustering keys today is short and monotonically increasing. Many will now consider a GUID as “short enough” so the remaining problem is the need … Continue reading More on Primary Keys and GUIDs – NewSequentialID()
I had the honour to spend the day in Sydney today with the Craig Bailey’s group at OzFox Lite. I was speaking about SQL Server 2005 and developers. It was really excellent to see the level of passion still present in the Visual FoxPro community. Thanks to all involved. Great to get to catch up with Adam Cogan and to see the level of support from Microsoft, particularly Andrew Coates.
Mitch created a CafePress store for CodeCampOz. I ordered 3 large mugs from it. I’m going to use them as swag / prizes at CodeCampOz. They’re here! Anyone planning on coming that hasn’t let us know, please do so to firstname.lastname@example.org soon.
One of the things I haven’t been keen on is the auto-deployment of source code when deploying an assembly from within a Visual Studio Database project. Each assembly that is sent to the database ends up with an entry in sys.assemblies. However, there can be multiple entries in sys.assembly_files for a single assembly. The binary of the assembly is one entry. Visual Studio also deploys the pdb file and the source code as additional files. Many see this as undesirable as it is another copy of the source code that isn’t in a source code control system ie: another copy … Continue reading Avoiding source code deployment with CLR projects
I spent some time at the University of Queensland way back in the 70’s. One of the things I remember most vividly was the physics pracs. In the pass-level class, they gave you the equipment, told you what you had to show and then told you step-by-step how to do it. In the honours-level class, however, they gave you the equipment, told you what you had to show but *didn’t* tell you how to do it. I’ve often mentioned to people that the prac I remember best was one where we went into the room and my recollection of what … Continue reading Measuring the speed of light with chocolate chips
Great to see our buddy (and fellow MVP) Kent Tegels now podcasting. Kent is well known for his Enjoy Every Sandwich blog. Check it out at: http://www.sqljunkies.com/WebLog/ktegels/articles/TOAE001.aspx.
The Product Feedback Center is a great initiative from Microsoft. I do struggle with some of the feedback/statuses though. Take this excerpt for example: Resolved as Not Reproduced by Microsoft on 2006-02-09 at 18:04:03 This issue has been fixed in upcoming SP1 release. I must be missing something as to how you can have an issue “not reproduced” yet “fixed”. How would you know you’d fixed it?
I ran into Scott Thornton at the VDNUG meeting in Melbourne a week or so ago. Scott was a student I had taught in a SQL Server class last year. He sent me a number of suggestions regarding enhancements to T-SQL he’d like to see, particularly ones he misses from Oracle. One that really caught my eye was on variable declarations. Currently, when you declare a variable in T-SQL, you need to provide the data type. Often you are picking a type which is compatible with an existing database column type. The problem is that if the column type in the … Continue reading TSQL Variable Definition based on Column Type
We had an internal discussion the other day about what makes a good data entry app. The vast majority of Windows applications I see are lousy for data entry purposes. The discussion was prompted by an Outlook-style app that had a scrolling data-entry form window. My current hit-list for good data entry apps is: 1. Must 100% be able to drive it with a keyboard (good test is to open up your app, put your mouse off to one side and try to work) 2. Tab order must be totally appropriate 3. Must be able to see everything without scrolling (see … Continue reading What Makes A Good Data Entry App