In the past, I was an unashamed fan of the Dilbert cartoon strip. In recent years though, the author (Scott Adams) has become a pretty divisive character in and around US politics and life in general. So I was intrigued by his book Win Bigly.
Nowadays, I keep hearing many people refer to him as being very pro-Trump and in the opinion stakes, that immediately cuts him off from half the US population, and the majority of the world.
2020-04-24
I remember when SQL Server 7 was released, it was such a big deal that you could easily grow and shrink database files, and indeed, you could have databases auto-grow, and, auto-shrink. That seemed pretty appealing if you wanted a low maintenance solution.
If only life was so simple.
One of the real challenges of auto-grow and auto-shrink is that you could have a database constantly changing size up and down, and that’s quite nasty, because each time you grow, you’re often using a different part of your file system.
2020-04-23
I mentioned in my last SDU_Tools post that I get really good feedback all the time from users of our free SDU Tools for developers and DBAs, but none more than when we have more and more date and time functions. Another pair of functions that we added in v19 that was recently shipped was StartOfWorkingWeek and EndOfWorkingWeek.
These are simple scalar functions that take a single date parameter, and return the date for the Monday and the Friday of the week that contains the date provided.
2020-04-22
I was working with another client recently, and they were changing the working IT environments for their staff. What struck me as odd was that they called the new environments the Modern ones.
Modern was actually the name of the environment. I’m sure they currently see the environments as the modern ones, but soon enough, it won’t be modern or new, and then the name looks really, really odd. In a few years’ time, they’ll have more recent ones, and it then gets tricky. Are they then the Even More Modern environments?
2020-04-21
While there are some other data types that I’d like to see added (for example a month data type), SQL Server has quite a rich set of date and time data types. Some of these were older data types and some new ones were introduced in SQL Server 2008.
An important one that was added in 2008 is the date data type. It’s just a date. It has no time associated with it and it covers year from 1 to 9999. It says that it’s based on the Gregorian calendar (i.e. our current calendar) but that’s a bit ambitious given it didn’t exist back in the year 1. But for the range of dates that we would work with, this will always work fine.
2020-04-20
I posted the other day that I don’t like negative ranges used for IDENTITY columns. I feel the same about the range of values for sequences, but when I mentioned that I prefer using sequences, a few friends asked why I preferred them.
Let me start by saying that I was surprised with the previous post, that the strong responses were about using natural vs surrogate keys. That had nothing to do with what I was discussing, and is an endless topic for another day. The issue was about when someone has already decided to use an IDENTITY or a sequence. And so, the same point applies to this discussion.
2020-04-17
This week, one of my colleagues was complaining that he had to do work on a database where every table had identity columns, and they all started at large negative values.
I remember some people promoting this years ago, to double the available numeric range for the data types, rather than just using positive values.
I didn’t like it when I saw it then, and I still don’t like it.
2020-04-16
I get really good feedback all the time from users of our free SDU Tools for developers and DBAs, but none more than when we have more and more date and time functions. One pair of functions that we added in v19 that was recently shipped was StartOfWeek and EndOfWeek.
These are simple scalar functions that take a single date parameter, and return the date for the start and end of the week that contains the date provided.
2020-04-15
I work with a number of clients in a variety of industries. I’m constantly amazed by the larger companies that simply do everything they can to avoid paying for things that they should be paying for.
I’ll give you two simple examples.
Many companies use TeamViewer. It’s easy to use and it works well for what it’s intended for. However I’d say that over 90% of the clients are using it as the free personal edition that says all over it for non-commercial use. I don’t get why companies that are turning over tens of millions of dollars, or who are managing billions of dollars of other people’s funds struggle to pay the correct licensing for basic utilities that they depend upon.
2020-04-14
In upcoming blog posts, I’ll show you how to change from one data type to another. I’ll show you how to use CONVERT to output date/time values to string formats.
If you look at the web page for CAST and CONVERT, you’ll notice there are many styles for conversion. For example, the US format is called 101 if it has a 4 digit century. But it’s called 1 if you only have a 2 digit century. Another very common one is the British/French one and that’s 103 for a 4 digit century or just 3 for a 2 digit century.
2020-04-13
