2015-11-22 - SQL

Structured Query Language (SQL) is a family of languages used for interacting with databases. Originally developed by IBM in the 1970 and standardized in 1986. SQL allows users to retrieve, create and modify data in a database. These queries are commonly made against a database engine that transforms the query into a series of actions that are required to complete the requested operation.

One of the best things about SQL is its query syntax. For simple operations these queries are very natural sounding which makes them easy to work with and understand. For example to get column1 and column2 from table1 where column1 is equal to 3 you would write this.

If you wanted the results to be ordered by coumn2 you would change it to.

This makes SQL a very approachable language. Simple operations are simple to perform and it’s only when you start wanting to do more complex things that the query start becoming more complicated.

The main reason SQL can have such simple syntax is because it’s a declarative language. This means that instead of telling the database engine what you want it to do you declare what you want to happen. It’s the database engine’s job to figure out how to do it. In a lot of languages doing something a bit weird feels like cheating, like you are doing something you shouldn’t be doing. In SQL it feels really good when you get a weird query to work exactly how you want it to be. I think this is largely because you aren’t saying what you want it to do so you don’t feel like you are forcing it to do something. Instead you are just finding the best way to describe what you want to happen.

SQL is a standard but is still somewhat un-standardized. All SQL languages are derived from the standard and have a lot in common but they aren’t all the same. The main differences come from the addition of procedural elements which are required for writing scripts that make decisions or perform operations multiple times. These procedural elements weren’t a part of the original standard so they were implemented differently by individual database systems. Microsoft uses Transact-SQL (T-SQL) for their SQL Server and Oracle uses Procedural Language/Structured Query Language (PL/SQL) for their database products. A standard for these procedural elements was created as SQL/Persistent Stored Modules (SQL/PSM) which some databases, such as MySQL, base their versions on. There are also slight syntax and data-type changes between implementations. This means that going from one database system to another does require some new knowledge. A lot of the basics are the same but you need to be aware of the actual implementation differences.

I personally like SQL. It’s a very quirkily language but in a lot of ways that makes it more enjoyable to work with. You can do basic things without a lot of knowledge but you need to really understand the particular database you are working with to do really advanced things well. The more time you put into understanding the quirks the more enduring the language is.

2015-11-08 - In IL: Introduction

The Common Intermediate Language (CIL), Microsoft Intermediate Language (MSIL) or usually just Intermediate Language (IL) is the language of the virtual machine component of the Common Language Infrastructure (CLI) standard. The .NET Framework and Mono are implantations of this standard.

The virtual machine is a lot like a computer but implemented in software. The virtual machine exectures IL programs by compiling it into native code. That native code is the language of the computer that the program is running on and represents the operations that it understands. The virtual machine also handles a lot of tasks that would normally need to be done by the program itself such as allocating and freeing memory. This simplifies programming and also provides isolation between the program and the actual computer.

IL provides an intermediate step between the higher level languages and the native code of the computer. It provides classes and functions which are common in high level languages while at the same time having a syntax and structure more like native code. This makes it familiar to people approaching it from the higher level language point of view and from the native code point of view allowing both sides to easily work with it. It also separates responsibilities because the language designers only need to focus on the IL compiler while the designers of the virtual machine implementation focus on translating the IL into native code. Any changes to the IL compiler benefit all implementations it supports. Any improvements to the virtual machine benefit all the languages that target it.

In this series of posts I want to investigate how languages that target the .NET Framework (C#, Visual Basic .NET) compile into IL code. The idea is to learn a little bit about IL, how the compiler works, and how .NET programs actually run. This information should be useful when designing or debugging programs and may even lead to some ideas about how to build a compiler.

Should be fun.

2015-10-18 - Character Encoding: Conclusion

I actually don’t spend a lot of time dealing with character encodings. It comes up a little bit when dealing with files but even then it’s just a matter of selecting the right encoding. So then why did I spend the time to investigate and write about character encodings?

Partially it’s because it’s a good thing to be aware of. It’s useful to understand what selecting the encoding is doing. Also the problem of how to store non-numeric data comes up a lot in programming and character encodings serve as a good example of that problem.

That being said the main thing is how it shows that the best solution to a problem can change over time. The designers of ASCII made the decision that data savings was more important that number of characters so they limited themselves to 7 bits. As time went on this situation changed; space became cheaper while the desire for more characters grew. This lead to Extended-ASCII and Unicode with their 8 bit, 16 bit, and 32 bit characters. Then things partially flipped around again. With the internet and data being sent all over the world space savings became important again but people didn’t want to lose their extra characters. This lead to the UTF formats that went for space savings under common circumstances at the cost of added complexity.

ASCII and Extended ASCII don’t fit current needs because the limited character set and need for code pages complicates sharing information around the world. Similarly UTF-8 wouldn’t have worked for early computers and Teletype machines because the variable width characters would have made it excessively complicated to implement on the hardware of the time.

I find problems that don’t have singular answers to be the most interesting. Problems with the specifics of the situation impact the requirements and multiple solutions can work simultaneously in different situations. All of these encodings are in use today. Some more than others but the introduction of newer encodings hasn’t destroyed those that came before it. One of the goals of UTF-8 was to be compatible with ASCII for this reason.

I’d also like to point out that this is not an exhaustive list. There are a bunch of other character encodings. Even within Unicode their are other transformations and variations of transformations. These are just the most common ones that I know of and have encountered.

2015-10-04 - Seriously Fun TV

I really like TV. I watch a fair bit of it, and from all that watching I’ve found that there’s a recipe to the shows that I enjoy. The best TV shows are those that start with a large scoop of drama and then add in a few spoonfuls of comedy on top of it.

I like stories and characters. Those are the things I find interesting about TV shows. So I tend to prefer dramas because they usually have better stories and characters. The problem is that shows that are pure drama tend to be very dark and depressing. They are just about bad things happening to the main characters and their trying to deal with things before the next bad things happen. I personally don’t like shows that make me depressed.

That’s where the comedy comes in. A little bit of silliness helps to break up the drama in the story and make the characters more likable. The comedy keeps things from being too depressing. The show can’t just be a pure comedy though because then you lose the stories and characters. They just become a series of jokes and as funny as they may be they aren’t really interesting.

So my favourite shows are those that have good stories and characters but aren’t too serious.

17 18 19 20 21